Citation
Methane-related authigenic carbonate formation on the Cascadia accretionary prism

Material Information

Title:
Methane-related authigenic carbonate formation on the Cascadia accretionary prism
Creator:
Bateman, Mary Lindsey
Publication Date:
Language:
English
Physical Description:
vii, 107 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Aragonite ( jstor )
Breccia ( jstor )
Calcite ( jstor )
Carbonates ( jstor )
Dolomite ( jstor )
Hydrates ( jstor )
Methane ( jstor )
Precipitation ( jstor )
Sediments ( jstor )
Sulfates ( jstor )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (M.S.)--University of Florida, 2002.
Bibliography:
Includes bibliographical references.
General Note:
Printout.
General Note:
Vita.
Statement of Responsibility:
by Mary Lindsey Bateman.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright; Bateman, Mary Lindsey. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
51947199 ( OCLC )
ocm51947199

Downloads

This item has the following downloads:


Full Text









METHANE-RELATED AUTHIGENIC CARBONATE FORMATION ON THE
CASCADIA ACCRETIONARY PRISM
















By

MARY LINDSEY BATEMAN


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2002













LD
1780 20 --,A32?
























Copyright 2002

by

Mary Lindsey Bateman

























For my parents and grandparents, whose support and love made this thesis possible













ACKNOWLEDGMENTS

I first want to thank Dr. John Jaeger for his constant input and help with the research and writing of this thesis. The other members of my committee, Drs. John Martin, Anthony Randazzo, and Kenneth Sassaman also have been invaluable for referrals and discussions. I would like to extend my appreciation to Kendall Fountain and Jason Curtis for their help running samples and for their useful suggestions.

I would also like to thank my parents and my grandparents, who have been my strongest support during my years at UF. Their guidance, patience, and unconditional love ultimately were the driving forces behind this thesis. Aleta Mitchell-Tapping and Timothy Gysan also played an integral part in supporting me and helping with proofreading and graphing. Thanks also to Marc Shook, Christy Gysan, Elena Miranda, and Mark Leidig, who were a valuable outlet for discussions during research and writing. Finally, I would like to thank all my close friends who have always given me support and love, and without whom my career would not be the same.














TABLE OF CONTENTS



ACKN OW LEDGM ENTS................................................................................................. iv

ABSTRACT ....................................................................................................................... vi

CHAPTER

1 INTRODUCTION ............................................................................................................ 1

2 STUDY AREA (SITE LOCATION) .............................................................................7

3 LITERATURE SURVEY (BACKGROUND INFORMATION) ................................. 9

4 M ETHODS ..................................................................................................................... 19

5 RESULTS....................................................................................................................... 22

6 DISCUSSION ................................................................................................................79

7 CONCLUSION S ........................................................................................................... 95

APPENDIX DATA FROM SAM PLES.......................................................................... 97

LIST OF REFERENCES ............................................................................................... 102

BIOGRAPH ICA L SKETCH ......................................................................................... 107


















V













Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

METHANE-RELATED AUTHIGENIC CARBONATE FORMATION ON THE CASCADIA ACCRETIONARY PRISM By

Mary Lindsey Bateman

December 2002

Chairman: John Jaeger
Major Department: Geological Sciences

Methane-related authigenic carbonate formation has been documented in the rock record as well as in modem oceans around the world. Knowledge about the unique conditions under which the carbonate forms is important to understanding fluid flow on continental margins and the fate of carbon. The examination of the processes influencing carbonate precipitation on Hydrate Ridge is the basis for a Masters thesis in geology. Hydrate Ridge, an elongated ridge on the accretionary prism located off the coast of Oregon, represents a small part of the subduction zone between the Juan de Fuca and North American plates. High organic carbon fluxes result in widespread methanogenesis in the prism sediments producing large stores of methane hydrates. Thrust faults cutting the accretionary prism serve as conduits for the methane to migrate upwards to the top of the seabed, where it is trapped as gas hydrates or is released into the water column.

Three carbonate mineralogies are present on Hydrate Ridge: aragonite, high Mg calcite, and dolomite. This thesis presents a hypothesis on the depth of origin of the





vii


different mineralogies based on a variety of influencing factors. Analyses performed on the samples to understand their origin included optical and scanning electron microscopy, X-ray diffraction, carbon and oxygen isotopic analyses, and determination of the total mass percent carbonate. These analyses were aimed at understanding the sedimentary conditions under which the carbonate precipitated, and included porosity, availability of oxygen and sulfate, and methane concentrations as either free gas or hydrates.

The primary factor controlling the carbonate mineralogy promoted in this study is the amount of dissolved sulfate in the formation waters. Aragonite precipitates at the seabed surface where sulfate concentrations are high. High Mg calcite preferentially precipitates where sulfate is depleted, in the upper sediment column at 4-10 cm depth. Dolomite precipitation is also inhibited by sulfate, and therefore its precipitation occurs in the sulfate-reduction zone in the upper couple of meters of the seabed, as shallow as 4 cm depth. The high concentration of dolomite present on Hydrate Ridge could be a result of recrystallization from high Mg calcite and precipitation in larger quantities in the past due to higher temperatures of migrating fluids (over 1000C), or a result of pockets of extremely depleted sulfate in the sediment column. Breccias are common on Hydrate Ridge, and can be composed of high Mg calcite and aragonite clasts. They are proposed to be the result of hydrate destabilization, which fractured carbonates that then were re-cemented. The presence of pockmarks on Hydrate Ridge is evidence of this destabilization.













CHAPTER 1
INTRODUCTION

Authigenic carbonates play an important role in our understanding of overall fluid flow through convergent continental margins. Tectonic convergence on these margins leads to dewatering of sediments accompanied by active fluid flow along fractures and faults. This fluid flow has received much attention for its contribution to global geochemical fluxes, especially the global carbon cycle (Aharon 1994, Carson et al. 1994, Kastner et al. 2000). Methane is commonly produced on convergent margins by the burial of organic material from both biogenic and thermogenic methanogenisis.

Methane within continental margin sediments can be found in three phases

according to local temperature and pressure. At sufficiently high temperatures (9.50C) and pressures deep in the sediments (20-2000 mbsf), methane is found as a free gas (Hyndman et al. 1995, Kvenvolden 1995). As pressures and temperatures drop with a rise toward the seafloor, methane gas binds with porewaters to form methane hydrates. The transition from free gas to gas hydrate deep in the sediments approximately coincides with the bottom simulating reflector (BSR). Methane gas can also migrate upward through faults and fractures and can mix with the seawater in the water column. Methane-oxidizing bacteria use free methane as an energy source, creating bicarbonate or CO2 as a byproduct of the methane oxidation reaction depending on the oxidant used (sulfate or oxygen, respectively; specific equations discussed in Chapter 3). The pore water in the shallow sediment column and the near-bottom seawater becomes saturated








with bicarbonate, creating a favorable environment for the precipitation of carbonates (Ritger et al. 1987, Suess et al. 1998).

Methane seeps and related authigenic carbonates occur around the world (in both deep and shallow water) including the Gulf of Mexico and North Sea, and off the coasts of Denmark, Florida, Cascadia, and Barbados (Jorgensen 1992, Kastner et al. 2000, Martin et al. 1996, Paull et al. 1992, Rad et al. 1996, Suess et al. 1998,). These methane seeps often occur in accretionary prisms associated with plate subduction, and therefore the presence of authigenic, methane-related carbonates may prove useful in identifying similar deposits in the rock record. Ancient suspected cold-seep carbonates have been recognized in the rock record across the world (Canada, Washington State, and the Apennines) (Beauchamp and Savard 1992, Cambell 1992, Terzi et al. 1994). Globally, methane-related authigenic carbonates are composed of surprisingly similar mineralogies that include aragonite, high Mg calcite, and dolomite, which all appear regardless of hydrate occurrence or water depth. This similarity in mineralogies suggests a simple, common set of ubiquitous processes control the precipitation of different mineralogies.

To better define the conditions in which methane-related carbonates form, a

systematic understanding of fluid flow and corresponding methane dynamics in prism sediments is necessary. One region where a basic understanding of these processes exists is on the Cascadia Margin. In a field site known as Hydrate Ridge (Figure 1-1) (Bohrmann et al. 1998), a series of studies were conducted to evaluate fluid flow along faults (Westbrook et al. 1994). Early work in this region by Kulm et al. (1986) recognized a connection between active methane seeps at the seafloor and the widespread occurrence of authigenic carbonates (Carson et al. 1994). Recent work on Hydrate Ridge








suggested a close relationship between gas hydrates and authigenic carbonate formation. The carbonates are concentrated where methane-saturated fluids seep from the seabed, and are hypothesized to form by the aforementioned reaction between methane and sulfate, mediated by microbial communities. Bohrmann et al. (1998) recovered carbonates within gas hydrates, and proposed carbonate precipitation in and around the hydrates in the upper 50 cm of the sediment surface. Yet, this connection cannot fully explain the global similarities in carbonate mineralogies (aragonite, high Mg calcite, and dolomite) and morphologies (chemoherms- discussed in Study Area section, crusts, slabs, boulders, and chimneys), especially in regions where gas hydrates are not present.

Because of the global similarity of carbonate mineralogies and morphologies on continental margins, the processes leading to precipitation of the three mineralogies must be equally common. Simply, mineralogy is influenced by the kinetics of precipitation, whereby a particular phase is favored. Under most porewater conditions at ocean bottom temperatures, Mg calcite is the favored form; but when even minor amounts of sulfate are present (greater than 5% of seawater value), aragonite is favored (Baker and Kastner 1981, Burton and Walter 1987) (Figure 1-2). Calcite is also inhibited by high concentrations of hydrated Mg2+ ions (present in ocean water), promoting aragonite precipitation instead of calcite (Ritger et al. 1987). In this setting on Hydrate Ridge, sulfate is most common in the seawater and shallow sediments, but is rapidly consumed at depth in the sediment column by methane oxidation. This system is likely common in methane-rich margin sediments.

The goal of this thesis is to investigate the formation of authigenic carbonates

related to methane seepage and possibly related to hydrate formation and dissassociation








on Hydrate Ridge by using carbonate samples and cores collected from the seabed by DSRV Alvin. More specifically, the hypothesis presented in this thesis is that aragonite is the favored form precipitated at or near the seabed where sulfate-rich waters are present. Other phases (high Mg calcite and dolomite) form deeper in the sediments (greater than 4 cm depth) where sulfate is minimized through methane oxidation. By using proxy indicators of the sediment depth of formation and availability of sulfate, this hypothesis was tested. These proxy indicators include porosity, availability of oxygen and sulfate, and methane concentrations as either free gas or hydrates. Porosity estimates were determined through petrologic examination of samples (optical and scanning electron microscope) and estimates of porosity replacement were determined by carbonate cement. Proxy evidence of the availability of oxidants was determined through petrologic examination of reduced or oxidized mineral phases (i.e. pyrite, glauconite, and hematite). Finally, the influence of deep methane hydrates over seawater was established through stable isotopic measurements of P13C and 8"0.


















44 40'N


44:
Ln'



44 35' N








0 KM 10

Figure 1-1. Study area of Hydrate Ridge, located off the coast of Oregon.
Carbonate Samples collected in this study came from the northern portion of
Hydrate Ridge. Pockmarks (not shown) are located on the east side of the
Ridge.

















3.0


2.5


5oC


a Aragonite a Calcite


n gOO


Figure 1-2. Precipitation rates of aragonite versus calcite at 50 C. Aragonite is slightly favored at these temperatures, as shown in this figure. However, the addition of sulfate inhibits the precipitation of calcite. (Modified from Burton, E., L. Walter. 1987. Relative Precipitation Rates of Aragonite and Mg Calcite from Seawater: Temperature or Carbonate Ion Control? Geology, v.15. 111114. Figure 1, pg.112)













CHAPTER 2
HYDRATE RIDGE STUDY AREA

Hydrate Ridge is an elongated anticline bisected by thrust faults associated with subduction. The study area is located 102 kilometers off the coast of Oregon (16 km from the Cascadia deformation front) on the accretionary prism, at which ODP borehole site 892B is located (Figure 1-1). High organic content in the sediment column has caused methane venting in this area for at least the past 21 to 24 ky (Kastner et al. 2000). Hydrate Ridge is named for the prolific hydrate formation in the prism sediments. Typical water temperatures on the sea floor of Hydrate Ridge are approximately 4.30C (Kastner et al. in press). Estimated sedimentation rate of hemipelagic and terrigenous material is 2 cm/ky, and sediments are generally comprised of interbedded glauconitic quartz sands and silty clays (Ritger et al. 1987, Schluter et al. 1998, Westbrook et al. 1994).

Extensive carbonates, often associated with fault expressions on the surface of the seabed, are scattered around Hydrate Ridge. Massive (3-5 meters high) aragonitic chemoherms, carbonate build-ups precipitating from chemical reactions between methane-rich fluids from depth and seawater, are outcropping at the seabed (Bohrmann et al. 1998). These chemoherms are frequently associated with fluid expulsion from faults, and can be identified through GLORIA side scan imagery (Carson et al. 1991). Chemoherms are often located at the top of the ridge, while muddy bottoms occur at the lower slopes of the ridge. Large boulder fields are located inside the pockmarks, and





8


occasionally on the lower slopes away from the ridge. Along with authigenic carbonates, methane-related clam communities, bacterial mats, and bubbles of methane gas confirm active venting on Hydrate Ridge (Westbrook et al. 1994).













CHAPTER 3
LITERATURE SURVEY (BACKGROUND INFORMATION)

The discovery of gas hydrates off the coast of Oregon was first noted by Kulm et al. (1986). Hydrates from Hydrate Ridge were then collected and described by Bohrmann et al. (1998). He noted the presence of carbonates near the surface on Hydrate Ridge, (700 meters) and bottom temperature (4oC). Kastner et al. (1998) noted floating hydrates and methane plumes in the water column over Hydrate Ridge caused by the significant amounts of gas hydrates expelled from the sediment column.

Sediment geochemistry of Hydrate Ridge was analyzed in ODP logs and results (1994). The logs documented high concentrations of H2S in the sediment column from near-surface to 60 meters depth. The results also concluded that bacterially-mediated methanogenisis occurs in the upper part of the sediment column (probably upper

2 meters). In the upper 20 meters, pore water Cl, Ca2+, Na, and alkalinity varied slightly with the dissassociation of the methane hydrates. Below 20 meters, decreases in Cl, Na , and Mg2+ concentrations and alkalinity are possibly the cause of advecting fluids (Westbrook et al. 1994). Schluter et al. (1998) sampled fluids from the borehole, and found that 02, NO3-, Si', H2S, SO42+, and Cl are all compositionally close to bottom water. Methane was considerably atypical, ranging between 0.2-3.5 mM, which is 106 times greater than seawater concentrations (Schulter et al. 1998). The sediments drilled from the borehole were classified as Pliocene terrigenous silty clays and clayey silts with some sand layers and Pliocene turbidites (Westbrook et al. 1994).








Mineralogy

Global observations of carbonates associated with methane consistently describe three mineralogies: aragonite, high Mg calcite, and dolomite (Jorgensen, 1990, Kastner et al. 2000, Martin et al. 1996, Paull et al. 1992, Rad et al.1996, Suess et al. 1999). Ritger et al. (1987) examined dredged and Alvin-collected samples from the lower Cascadia continental slope in thin section and SEM for mineralogies and crystal morphologies, which included radiating aragonite crystals and detritus-rich micritic calcite cement. Ritger et al. (1987) also found that all high Mg calcite and dolomite samples were comprised of microcrystalline cement. They found cubic or framboidal pyrite in every thin section they analyzed. In thin section, aragonite primarily formed as "elongate, radiating crystals" (Ritger et al. 1987).

Kulm and Suess (1990) also observed samples from the Cascadia continental slope that contain patchy pure aragonite and a micritic cement mixture of silt and clay. They found carbonates that were comprised of well-sorted glauconitic grains, with some pyrite. This dark glauconitic matrix was cemented by a combination of aragonite and high magnesium calcite, or by pure aragonite (Kulm and Suess 1990). Bohrmann et al. (1998) noted the pure botryoidal and isopachous aragonite in samples collected from Hydrate Ridge, and hypothesized that the crystals grow in gas hydrate cavities. The aragonite needles were 2-10 p.m thick, with botryoidal radii of 3-15 mm or layers of 40 to 600 p.m thick (Bohrmann et al. 1998).

Sample and Reid (1998) described high Mg calcite and dolomite samples collected by Alvin from the Cascadia accretionary wedge with multiple veins, and occasionally floating detrital grains located in the veins. They also noted minor amounts (0-5%) of








unbroken foraminifera tests, which could indicate a shallow cementation process (Sample and Reid 1998).

Utilizing X-ray diffraction (XRD) to determine mineralogy, Ritger et al. (1987) found dominantly high magnesium calcite cement which contained 6-23 mol% MgCO3, with an average of 12 mol% MgCO3. This average was much higher than normal deepwater calcites, which contained less than 4 mol% MgCO3. Ritger et al. suggested this could point to a precipitation at higher temperatures (Ritger et al. 1987). Greinert et al. (2001) used XRD analysis to classify carbonates from Hydrate Ridge into different phases based on amounts of magnesium: aragonite, low Mg calcite (less than 8 Mol% MgCO3), high magnesium calcite (8-20 Mol% MgCO3), protodolomite (30-40 Mol% MgCO3), and dolomite (40-55 Mol% MgCO3). In a study of gas hydrates in the sediment, Bohrmann et al. (1998) recognized two types of carbonates directly related to shallow methane hydrate deposits, including aragonite (most common) and high Mg calcite (14-19 Mol% MgCO3). Kopf et al. (1995) found that 80% of their borehole, Resolution-collected samples from the Cascadia accretionary wedge were composed of high magnesium calcite, while the remaining 20% were aragonite and dolomite. Sample and Reid (1998) used XRD to identify minor constituents in fine-grained carbonate samples. Secondary minerals found in XRD analysis of authigenic carbonate samples from the Cascadia margin included quartz, plagioclase, clay minerals, pyrite, and glauconite (Sample and Reid 1998).

Ritger et al. (1987) analyzed their Cascadia samples for the total mass percent

carbonate through an acid-leach weight-loss procedure, and the samples grouped into two categories. The first were low-carbonate samples, with less than 20% CaCO3, while the second had much higher amounts of carbonate, between 25% and 90% CaCO3. Sample








and Kopf (1995) analyzed their high Mg calcite samples from off the coast of Oregon for total percent carbonate, and found a range between 1% and 10% CaCO3. A second area analyzed had higher carbonate content (up to 25% CaCO3). Total percent carbonate did not have any clear correlation with pore fluid chemistry, or Mg/Ca ratios (Sample and Kopf 1995).

Morphologies

On a global basis, the morphologies of authigenic carbonates are also similar. Carbonate chemoherms, crusts, slabs, boulders, and chimneys are common across the globe, and Hydrate Ridge is no exception (Kastner et al. 2000, Jorgensen 1990, Martin et al. 1996, Paull et al. 1992, Rad et al. 1996, Suess et al. 1999). Figures 3-6 show examples of chemoherms, crusts, slabs, and boulders derived through Alvin dives on Hydrate Ridge and in hand samples. Ritger et al. (1987) found Hydrate Ridge breccias that included mudstones, sandstones, and conglomerates and were often cemented by magnesian calcite. Greinert et al. (2000) petrographically categorized various authigenic carbonate phases from Hydrate Ridge as homogeneous mudstones, tectonized mudstones, bioturbated mudstones, mudclast breccias, intraformational breccias (clasts of intraclasts, extraclasts, shells, and mudclasts), cemented bioturbation trails, and gas hydrate-associated carbonates. Mudstones in general were cemented by dolomite to protodolomite and were rich in pyrite. The mudclast breccias were grain to matrix supported, with areas of centimeter-thick bands of pure radial aragonite crystals. The irregularly shaped intraformational breccias included intraclasts (sometimes glauconitic), extraclasts, bioturbation casts, shells, and mudclasts. Both the clasts and the matrix contained pyrite, and the cement was primarily aragonite needles. Greinert et al. (2000) divided








gas-hydrate related carbonates into aragonitic collapse breccias and elongate pores (gas hydrate relic). The collapse breccias were grain supported and cemented by boytryoidal aragonite. The elongate-pore carbonates exhibited distinctive megapores, which resemble gas hydrate layers (Greinert et al. 2000).

Factors Influencing Carbonate Precipitation

Extensive research has been published on the factors influencing the precipitation of authigenic methane-related cold-water carbonates, particularly from the accretionary prism off Oregon. Ritger et al. (1987) presented hypotheses on cementation processes and the conditions that lead to carbonate precipitation based on samples collected from the lower continental slope of Cascadia. They believed that cementation was induced by increased alkalinity associated with sulfate reduction, abundant availability of Mg2+ and Ca2+ ions from seawater, and decreased CO2 solubility due to a pressure decrease. In anoxic conditions, methane is oxidized and sulfate reduced by the reaction: CH4 + SO42 -- HS- + HCO3 + H20 (Ritger et al. 1987)

At the oxic seabed surface, the methane is oxidized by the reaction:

CH4 + 202 -4 CO2 + 2H20 (Ritger et al. 1987)

The calcium that combines with the CO32- to enable precipitation of carbonates is from seawater. Ritger et al. (1987) believed that the aragonite was younger than the calcite because aragonite needles crystallized on the calcite cement.

One important question that many studies have addressed is how so many different types of carbonate (aragonite, high magnesium calcite, and dolomite) are found in the same area. To answer this question, an understanding the conditions under which aragonite and calcite precipitate is necessary. Burton and Walter (1987) demonstrated that








the primary controls on carbonate phase precipitated are temperature and carbonate saturation state of seawater (Figure 1-2). At 5C (the average water temperature at Hydrate Ridge), rates of calcite precipitation are equivalent to those of aragonite regardless of saturation state, but as temperature increases, aragonite is the favored precipitate. However, an important control on the preferred phase is dissolved sulfate (Burton and Walter 1987). In the zone of sulfate reduction, high magnesium calcite is the favored phase, while in the sulfate-rich pore waters near the surface, aragonite is the favored form. In laboratory experiments, the presence of dissolved sulfate strongly inhibits the precipitation of calcite and dolomite. The reduction of SO42 assists dolomitization by removing the SO42- inhibitor, increasing the alkalinity, and producing NH4+ that can exchange with Mg2+ and free the magnesium for dolomitization (Baker and Kastner 1981).

Morse et al. (1997) studied the importance of temperature and Mg:Ca ratio in pore water on precipitation of calcite and aragonite. They found that both factors influenced the type of precipitation, but that minor temperature changes dramatically effected the Mg:Ca ratios. Higher temperatures (+60C) and normal seawater Mg:Ca ratios (5:1) favored aragonite precipitation, while low temperatures (<60C) and a Mg:Ca ratio less than 1:5 favored calcite formation. Therefore, Morse et al. also suggested that paleoclimate/temperature could greatly effect carbonate phases precipitated, and should be taken into account. However, wide changes in temperatures are not likely to occur on Hydrate Ridge due to the depth and near-zero temperatures.

Dolomite formation in methane-seep environments is more difficult to explain than aragonite and high Mg calcite. Some authors argue for direct dolomite precipitation, while other argue for a recrystallization of calcite to dolomite over time. Baker and








Kastner (1981) suggested the controls on dolomitization include not just the Mg/Ca ratio (as previously believed), but also the concentration of sulfate in the formational waters. Teal et al. (2000) also emphasized the importance of microbial sulfate reduction, along with methanogenisis. They suggested that dolomite precipitation was bacterially mediated at normal seawater salinity in anoxic conditions promoted by methanogenisis. They also found that the dolomitization was not obviously correlated with depth, location, sediment texture, porosity, or mol%MgCO3. Dolomitization in this model would be accompanied by authigenic pyrite precipitation. They also supported a recrystallization theory (calcite to dolomite) through burial diagenisis under some conditions on the seafloor (Teal et al. 2000).

A paper by Baker and Bums (1985) covered the occurrence of dolomite on continental margins. In areas similar to Hydrate Ridge, they concluded that primary controls on dolomitization were the availability of calcium and the concentration of sulfate. Similar to other published papers, they promoted a theory of dolomitization in the upper 20-30 meters of the seabed in the zone of sulfate reduction. This depth of precipitation is also supported by the availability of magnesium, which decreases more slowly with depth than calcium. Therefore, dolomite can precipitate at sediment depths greater than calcite. The sources for magnesium are varied, and may include calcite, clays, and diffusion from seawater. Finally, Baker and Bums also proposed the primary input of magnesium was from diffusion, and therefore the dolomite must form in the upper few tens of meters of the seabed (in the zone of sulfate reduction) (Baker and Bums, 1985).








Stable Isotopes

Oxygen and carbon isotopic analyses have been powerful tools in the study of authigenic carbonates because 813C and 8180O values can reflect methane-related carbon source, and can constrain porewater compositions and temperature at formation, respectively. 813C values can reflect the methane source (biogenic or thermogenic), and the influence of seawater during precipitation. Typical 813C values for seawater are around 0%o PDB, while biogenic sources are highly depleted, recording 813C values of

-75%o to -90%o PDB, and thermogenic sources record 813C values of -30%o to -50%o PDB. However, mixing of seawater and biogenic methane can increase the 813C values to the thermogenic range. Kvenvolden (1995) suggested that when methane was the dominant gas hydrate, then the 813C values were most likely due to microbial methane. The 8 13C values of carbonates from Hydrate Ridge were analyzed for the influence of hydrates, methane gas, and seawater during precipitation (Kastner et al. 2000, Suess et al. 1999).

8 I8O values of carbonates are controlled primarily by mineralogy, temperature, and the oxygen isotopic values of the water from which the carbonate precipitated. It is important to note here that oxygen isotopic values can be effected by climatic changes during the Cenozoic (through temperature changes), but these changes are not significant on Hydrate Ridge in the past 20 kyrs (Hudson and Anderson 1989). Most 818O values from Cascadia are primarily affected by the isotopically-heavy waters (with respect to seawater) which are produced by methane hydrates (+0.1%o to +7.9%o). Methane-rich waters are generally isotopically heavy, and will affect the ~8 8O values for precipitating carbonate (Matsumoto 1989).








Ritger et al. separated their isotopic analyses into low carbonate and high carbonate samples. The low carbonate samples (less than 20% CaCO3) had carbon isotopic values of -0% PDB, but had unusually low 6180 values (-1.6%o to -13%o PDB). The high carbonate (over 25% CaCO3) samples reflected a methane influence as seen in 813C values that ranged from -66.7%o to -34.9%o PDB. Positive 8s180 values (+2.78%o to +8.27%o PDB) in the high carbonate samples reflected "a more complex signal than just temperature of formation" (Ritger et al. 1987). Ritger et al. believed in a biogenic methane source because the measured carbon isotopic values were fairly depleted in 13C. Sample et al. measured a range of -1%o to -25%o PDB for carbon isotopes from samples collected by Alvin along the frontal thrust in Cascadia, which were assumed to reflect a thermogenic source. The measured oxygen isotope values were extremely depleted (-4%o to -25%o PDB), indicating high temperatures (1000C) during precipitation (Sample et al. 1993). The isotope data for the carbonate cements trended along a line from depleted in 6180 and average in 8'3C to enriched in 180O and depleted in 813C. Sample et al. hypothesized that the isotope values were a direct result of warm, deep fluids (1000C) migrating up through faults and precipitating carbonates at shallow depths (Sample et al. 1993).

Greinert et al. (2000) reported the widest range for carbon isotopic data occurred in chemoherm samples (+26.2%o PDB to -56%o PDB). These samples were then divided into six groups based on lithology, minerology, and carbon isotopic data (see previous discussion for lithologies). The dolomite mudstones comprised the most positive 813C values, 10%o to 26%o PDB. Greinert et al. (2000) proposed these positive values were a result of the reduction of CO2 in methanogenic zones. A second group, consisting of








mudclast breccias, mudstones, and cemented bioturbation trails, had carbon and oxygen values ranging from, -34.6%o to -46.7%o PDB and 6.3%o to 7.5%o PDB respectively. The high magnesium calcites from the next group had 813C values up to -29.2%o PDB and 180'O values around 4.3%o PDB. The majority of the samples fell into the last category, composed primarily of intraformational breccias and mudclasts. Their carbon isotopic values ranged between -38.2%o and -55.2%o PDB. Aragonite from this group had 8I80 values between 0%o and -0.3%o PDB (in equilibrium with ambient seawater), while the high magnesium calcite 180'O values fell between 4.9%o and 7.3%o PDB (in equilibrium with isotopically heavier water). They concluded that high magnesium calcite precipitated under different conditions than the aragonite, and were more influenced by gas hydrates reflected in the positive 8180 values (Greinert et al. 2000).

Bohrmann et al. (1998) found two isotopically distinct end members in a study of carbonates collected from Hydrate Ridge. These results, which are similar to Greinert et al. (2000), separated a hydrate-influenced high magnesium calcite and an ambient-porewater-influenced aragonite. The recorded carbon isotope values for the high magnesium calcite and aragonite were between -40.6%o and -54.2%o PDB, reflecting a methane influence in both phases. The aragonite oxygen isotopes averaged 3.7%o +/- 0.1%o PDB, while the magnesium calcite averaged 4.9%o +/- 0.1 %o PDB. The aragonite oxygen isotope value differed from SMOW by only 0.1%o, while the high magnesium calcite differed by 1%o (Bohrmann et al. 1998).













CHAPTER 4
METHODS

Collection of samples and push cores took place over a period of three summers in 1998, 1999, and 2000. DSRV Alvin collected carbonate hand samples and push cores, which were then returned to Lehigh University and the University of Florida for analyses. At Lehigh, cobble- and boulder-sized samples were cut with a rock saw into hand-sized specimens for thin section preparation and to study macroscopic texture and mineralogy (Figures 7-11). A total of 59 thin sections were produced by Texas Petrographics. These thin sections were analyzed and photographed through the use of a Zeiss photomicroscope III, primarily at x2.5 and x10 settings.

X-ray diffraction analysis was run on 20 samples at the University of Florida, all from push cores taken by Alvin. The cores ranged in length (10 cm and 18 cm), and were subsampled in two centimeter intervals. The intervals were then sieved into 4 size fractions, <-10, -10---1.750, -1.750---2.750, >-2.750. The largest fraction,

>-2.750, was photographed and then ground to 200g. Some of the clasts had obvious zoning, and three were drilled for zones (6 samples). All two-centimeter-interval samples (14 samples) and zoned clasts (6 samples) from the cores were run on a Phillips APD 3600 XRG 3100 generator at 45kV/30mA. They were run with a step size of .020, count time of 2 seconds, and 20 (peak angle range) of 24-38. The samples were mixed with an internal standard, corundum, at approximately 50/50 mass. Data were collected and interpreted using Jade 3.1+ software. XRD was also run at Lehigh University on 44








samples drilled from carbonate hand specimens. The generator and settings were the same as above. High Mg calcite samples were rerun (18 samples) to identify multiple peaks in the calcite-dolomite range (29.50-310 20), and qualitatively distinguish high Mg calcite from calcite and dolomite. The position of the peak in the calcite-dolomite range determined the mol% MgCO3, based on Goldsmith et al. (1961). Poorly ordered dolomite peak was assumed to lie at 30.250 20. The software used at Lehigh was Phillips APD1700.

Total mass percent carbonate was calculated using a model 5011 CO2 coulometer at the University of Florida. The software used to analyze and interpret the percent carbon was WinCarb. Thirty-seven samples were drilled from carbonate hand samples using a Dremel hand drill, and were weighed and analyzed. The samples were chosen from a variety of different lithologies, colors, clasts, and mineralogies. Some transects along a hand sample were drilled in order to study changes from rims to interiors. Clasts sieved from the push cores were also selected to be analyzed for total mass percent carbonate. Five clasts (>-1.750) from two intervals, 2-4 cm and 8-10 cm, were chosen as a representative sample and photographed. These were then ground, weighed, and analyzed. The percent carbon calculated by the coulometer was then multiplied by the known standard value of 8.33 to find total percent carbonate. The total weight percent carbonate was then converted to volume percent carbonate through the equation: Vol % Carbonate = (wt.% carb./2.90g cm-3)/((wt.% carb./2.90g cm-3)+(wt.% noncarb./2.65 g cmn 3)) x 100 (after Scamman 1981) Carbon and oxygen isotopic values were measured at both the University of

Florida and Lehigh University. The samples were ground by hand to a fine powder, and








then weighed. The isotopic values were measured by a Kiel III Carbonate Prep Device coupled to a Finnagan-MAT 252 isotope ratio mass spectrometer located at the University of Florida. A total of 101 samples were run at Florida, and 19 at Lehigh. XRD samples from the push cores were not run for isotopic values because the mixed mineralogy found by the XRD analysis would have greatly complicated the interpretation of the isotopic results. Each sample was evaluated for the IO8 value of formation waters given a temperature of 3.90C using standard equations for their respective mineralogies (aragonite and calcite) through the following equations: for aragonite (PDB) AA-B= (3.52(sqrt(3.522-(4*0.03*(19-T)))/(2*0.03) (Grossman and Ku 1981), and for Mg-calcite (PDB) AA-B= 2.78(106T-2)-2.89+0.06 mol%MgCO3 (Tarutani et al. 1969)

Scanning electron microscope analysis (SEM) was performed on six samples that represented the mineralogies present on Hydrate Ridge. The samples were trimmed to the appropriate size, sanded on their base to provide a flat surface (tops were left with natural texture), and then mounted using silver glue to stubs. The samples were then coated with iron and analyzed by a Joel JSM 6400 scanning microscope. Photographs were digitally recorded as 200 dp TIFF images. Individual grain mineralogies of SEM samples were determined through EDS (energy dispersive spectrometry), using the presence of Mg, Ca, C, and O peaks as an indicator of calcite or dolomite, and Sr, Ca, C, and O peaks as an indicator of aragonite.













CHAPTER 5
RESULTS

Previous field studies of Hydrate Ridge over the past 15 years revealed that authigenic carbonates occurred in several different morphologies which have been discussed previously (Figures 5-1 through 5-9). Chemoherms are large, porous mounds of carbonate (typically acicular aragonite) which are located at and above the seabed (Sample et al. 1993). The crusts are thin (1-2 cm) outcrops of carbonates jutting from the seabed, whereas the slabs are similar but thicker, (2-4 cm) less porous, and often lie in stacks. Crusts are generally composed of aragonite, while slabs are typically high-Mg calcite (see below).

Mineralogies

The mineralogies of samples were measured with XRD analysis (Table A-1). The XRD was performed on sections of 20 hand samples to identify mineralogy. The hand samples measured included aragonitic chemoherm samples, dolomitic samples, high magnesium calcite samples, and breccia samples. In these hand samples, multiple carbonate mineralogies did not occur in any one sample except for in the breccias (see figure 5-10 for example of XRD analyses). The different carbonates of chemoherm samples (Figure 5-5) were analyzed by XRD, and found to be aragonite and quartz (Table A-1). Ten samples from crusts (Figure 5-6) were tested with XRD and found also to contain aragonite and quartz (Table A-1). Carbonate slabs (Figure 5-7) collected from the seabed surface of Hydrate Ridge had been identified in the literature as containing high








Mg calcite. This observation was confirmed through XRD analysis of slabs from Hydrate Ridge (Appendix Table 1). Slab samples were composed of high Mg calcite, containing approximately 6 to 11 mol%Mg. The boulder samples were proto-dolomite and dolomite, and difficult to distinguish in XRD from very high Mg calcite (Figure 5-8, Appendix Table 1). Almost all dolomitic samples were disordered, with 39-42 mol%Mg. As stated in the methods section, this study considered any peak reflecting a spacing over 3.25 angstroms (30.250 20) to be highly disordered dolomite, and therefore these samples fell under protodolomite category (not high Mg calcite).

The breccias were more complicated than the other carbonates collected at Hydrate Ridge (Figure 5-9). They contained multiple mineralogies and lithologies in their clasts and cements. From XRD analysis, clasts consisted of feldspar and quartz, pure calcite, high Mg calcite, dolomite, and, less commonly, aragonite. However, aragonite was the dominant cement (Figure 5-9) seen in the breccias as determined by XRD. Some of the breccias had very similar clasts with the same minerology, but others had a wide variety of clast sizes and mineralogies. The high Mg calcite clasts ranged between 9 and 18 mol%MgCO3.

Finally, XRD was run on the largest, authigenic, hand-picked clasts from the bulk sediments sieved in the two push cores (sieved size -2.750). One sample from each two-centimeter interval was analyzed by XRD, and results can be seen in Figures 5-11 and 5-12. The first core analyzed (B20) was 10 cm long, and every interval has mixed mineralogies. Aragonite was present in all intervals except for 6-8 cm. High Mg calcite was also present in all intervals, but the peaks were wide for all intervals, implying a range of mol%MgCO3 in the calcite. Pure calcite was present only in the 4-6 cm interval.








Dolomite was also present in all intervals, and was all proto-dolomite, except for a double peak in the 4-6 cm interval of 50% near-stoichiometric dolomite (310 20) and 50% proto-dolomite. The dolomite became more disordered with depth in the core, lying closer to 300 20.

The next core (CJ4) was longer, at 18 cm (Figure 5-12). Aragonite was present in the top 14 centimeters, below which it disappeared. High Mg calcite was not present in the top two centimeters or the bottom 6 cm, but was present from 2 to 12 cm depth. The high-Mg calcite was concentrated in the center of the core. A small peak of low Mg calcite was noted in the upper two centimeters of CJ4. Dolomite was not present in the top two centimeters. The dolomite was primarily disordered, especially in the 6-8 cm interval and was more disordered in the center of the core. Both cores analyzed contained a detrital quartz and feldspar component as seen in XRD.

Due to the mixed mineralogy results from XRD analysis of the cores, some of the other large clasts from the same intervals in the cores as the clasts analyzed in XRD were drilled from brecciated sections or concentric zonations (examples can be seen in Figures 5-13 and 5-14). Two clasts with zones were chosen from CJ4 interval 12-14 cm. One of the clasts analyzed resulted in no carbonate found by XRD in either zones. Only a quartz peak was present in the diffractogram for both zones, so the clast apparently was comprised entirely of detrital material. The second clast from CJ4 12-14 cm had an interior zone consisting of dolomite with minor amounts of quartz. The exterior was also primarily dolomite, but had a peak of low Mg calcite. The final clast analyzed was taken from CJ4 interval 14-16 cm. The interior and exterior zones were composed of dolomite and minor quartz, and no other types of carbonate appeared to be present.








Porosity

The porosity of the carbonates collected from Hydrate Ridge was analyzed in thin section, SEM, and as total percent carbonate assuming complete filling of the pore spaces by authigenic carbonate. The mass percent calcium carbonate was calculated for many of the same samples also run for oxygen and carbon isotopes (Figure 5-15). Chemoherm samples had an average of 73% calcium carbonate, while the aragonitic crust samples had the highest average of 82% CaCO3. High Mg calcite samples had an average of 74% CaCO3, and dolomitic samples averaged at 61% CaCO3. Aragonitic samples had a much smaller range of percent carbonate values with a standard deviation of 9% than the high Mg calcite (standard deviation of 14%) or the dolomite (standard deviation of 28%). All percent carbonate values are listed in Table A-2. Total percent carbonate by weight was also converted to volume percent carbonate, which estimates the minimum initial porosity in the sediments before precipitation and replacement of pore volume with carbonate cement. This calculation implied the same trends in porosity noted above, with the highest porosities in aragonite (55%-90%) and the lowest in dolomite samples (42%-84%) (Table 5-1).

Volume percent calcium carbonate was also run on five small clasts from two

intervals in two cores, B20 and CJ4 (Figure 5-16). Interval 0-2 cm in B20 had an average of 75.6% CaCO3, which then dropped to near-zero averages in the 2-4 cm and 4-6 cm intervals. The 6-8 cm and 8-10 cm intervals increased back to 70-80%. Core CJ4 intervals primarily ranged between 65 and 83, with three exceptions. Intervals 6-8 cm and 10-12 cm had lower volume percent CaCO3 values at 46% and 51% respectively, while interval 14-16 cm dropped to 17%.








Finally, larger clasts from every interval in each core were analyzed for total

percent calcium carbonate. The percent carbonate for core B20 was fairly consistent, with a range between 71% and 78% CaCO3, except for interval 2-4 cm which had almost no carbonate (0.1% CaCO3). The second core, CJ4, had an average of 63% CaCO3, with a range from 18% to 84% CaCO3. No significant trends with depth were detected in the percent carbonate.

Porosity can also be determined qualitatively through thin section and SEM analyses by visually approximating pore space. Petrologic examination of carbonate samples from Cascadia revealed great variability in fabric and can be seen in Figures 19-42. The chemoherm samples had two distinctive colors in the matrix (dark grey and light tan- Figure 5-5), as well as white aragonite needles that grew into pore spaces. The chemoherm samples had the highest percentage of void spaces of all samples studied. These void spaces contained elongated acicular aragonite that often interlaced (Figures 5-17, 5-18, and 5-19). Banding in the aragonite seemed to be a dissolution feature, where the needles had dissolved and then precipitation resumed (Figures 5-20 and 5-21). The needles tended to grow in fan shapes "rooted" in the darker micritic matrix; the boundary between the needles and the dark matrix was pronounced (Figure 5-17). Benthic foraminifera were plentiful in the chemoherm samples, and were usually whole.

Aragonitic samples from crusts collected from the seafloor had a similar structure to chemoherm aragonite samples, but contained fewer, shorter aragonite needles and more micritic aragonite than the chemoherm samples in SEM analysis (Figure 5-22). Clamshells were often incorporated into the matrix (Figure 5-23), and dissolution features on the shells were present. These samples also had high amounts of porosity. Whole








forams and glauconite were common. The matrix was composed of distinct dark grey and light tan sections (Figure 5-24) which were the same basic mineralogy based on XRD analysis. Clay and detrital material was common in both colored matrixes, and could be seen in SEM (Figure 5-25). The boundary between these different matrixes varied between both distinct and diffuse, depending on the sample. Porosity decreased in the grey matrix when compared to the tan. Some structures such as veins (some infilled with aragonite) and voids that were parallel to shells were present in the crusts. One thin section had graded beds of clasts (similar to Figure 5-30).

Samples from a high magnesium carbonate slab were analyzed in thin section and SEM (Figures 5-26 through 5-29). These samples were much less porous, with no acicular aragonite present (Figures 5-26 and 5-27). The slab matrix in hand sample was a homogenous, medium to well indurated, micritic calcite (light grey in color), and had a thin darker-colored rind. The high magnesium calcite was all micritic, and therefore was not easily analyzed in thin section (Figure 5-26). These samples had much more detrital material present than in the aragonite samples (Figures 5-28, 5-29, and 5-30). Forams were present in all the thin sections, usually with a 50/50 mix of fragmented and whole. One hand sample had only fragmented forams.

Dolomite samples were collected from the surface in push cores and as boulders, often occurring in boulder fields. Their matrix was homogenous, typically light grey colored (similar to the high Mg samples), and highly indurated. As stated in the methods section, this study considered any 20 peak over 30.250 20 to be highly disordered dolomite, and therefore these samples fell under the proto-dolomite category. The dolomite samples had very low porosity (Figure 5-31 through 5-33). Fragmented forams








were present in every thin section. Subparallel veins similar to the ones in the high Mg calcite (Figure 5-30) were present in these dolomite samples, and they were occasionally infilled with acicular aragonite. The veins with aragonite tended to be located near the edges of the samples, possibly indicating secondary precipitation. In SEM images, both veins and micritic dolomite looked identical, with similar mineralogy and grain size in both (Figure 5-33). These dolomite samples had larger forams compared to previously described thin sections of calcite and aragonite.

Breccias are common on Hydrate Ridge, and twelve representative breccias were chosen for thin sectioning (Figures 5-34 through 5-36). Only two of the breccias analyzed contained more than a few forams, and those were primarily comprised of whole, large forams (Figure 5-36). These samples had various percentages of porosity, ranging generally from 10% to 60%, determined by estimation in thin section. Large fragments of bivalves (15 cm in length) were common in breccias, and the cement between clasts was typically acicular aragonite. As stated previously, the breccias were a mixture of mineralogically-similar clasts and distinct, mixed-mineralogy clasts, typically matrixsupported. Some clast boundaries were diffuse, while others were sharp. The clasts were subrounded to subangular, a variety of sizes, and were a mix of detrital and carbonate. Structure was present in the breccias in the form of veins and elongated pores seen in the clasts.

Oxidant Availability as Proxy for Depth in Sediments

Oxidant availability in the carbonate samples from Hydrate Ridge also show wide variability. The presence of pyrite and glauconite indicate reducing conditions, while hematite staining implies oxidizing conditions. The chemoherm samples contained pyrite,








generally rounded (framboidal) in shape, and no visually-detected hematite staining present in any of the thin sections. Noticeably little glauconite (seen in thin section as clusters of nodules) was present in the thin sections. In crust samples, framboidal pyrite was common, and hematite staining was absent.

The samples from high Mg calcite slabs contained pyrite that was present in larger amounts than in the aragonitic samples, primarily occurring as infilling veins or voids in the matrix (Figure 5-37), but also as framboidal clusters. A small amount of the pyrite and surrounding areas were stained with hematite in one sample, while the other had no evidence of staining. Very little glauconite was present in any of the thin sections. Structures such as parallel veins and cracks were common in these samples (Figure 5-38), with glauconite infilling veins in one thin section. Dissolution rinds were seen in many samples in thin sections (Figure 5-39).

Dolomite had framboidal and infilling pyrite present in thin section and SEM (Figure 5-40). The framboidal pyrite was larger than any seen in previous samples (>5 mm). Hematite staining was common in these samples, and was concentrated near veins or cracks. The presence of glauconite was rare in these thin sections. The samples also commonly contained a dark red weathering rind (2-4 mm thick), possibly composed of iron and manganese.

Every breccia had small amounts of pyrite present. The pyrite was present as both framboidal clusters and within veins, often in the same thin section. The amount of hematite staining varied with individual breccias, ranging from no stains to stains commonly occurring particularly concentrated in veins. The amount of glauconite was








highly varied in these thin sections. High amounts of glauconite and the presence of forams tended to correlate well in the breccias.


Stable Isotopic Evaluation

Stable isotope analysis was run on all hand samples used in thin section and XRD but not core samples (see Appendix Table 1). 813C values for chemoherm samples ranged between -40.9%o and -48.4%o PDB. The I180 values for the chemoherm samples plotted between 3.8%o and 4.6%o PDB. The aragonitic crust samples were analyzed, and the carbon isotope ranged between -41.9%o and -49.9%o PDB (Figures 5-41 and 5-42). Four unusually high points also occurred in the carbon isotope data for these samples, falling between -9.8%o and -0.5%o PDB (Figure 5-42). The oxygen isotope values for aragonitic crusts plotted between 4.1%o and 5.1%o PDB, with no significant outliers.

High Mg calcite 813C values ranged between -52.0%o and -30.0%o PDB, while the oxygen isotopic values were all positive (Figure 5-43). For two high Mg calcite slabs, samples from microdrill transects were analyzed (Figure 5-44), and the 8 13C values ranged between -30.5%o and -44.8%o PDB. The oxygen isotope values fell within 3.7%o and

5.4%o PDB. Figure 5-44 transects depicts more depleted 813C values towards the bottom of the slab. The oxygen values have no clear trends along the transect.

Dolomitic samples were analyzed, and the carbon isotope values ranged between

-34.1%o and -39.6%o PDB (Figure 5-45). One 813C outlier of +4% was noted in one of the dolomitic samples. The 8180 values for the dolomite fell within 5.5%o and 7.2%o PDB, except for one outlier of -41.0%o PDB (Figure 5-45). These two carbon and oxygen isotopic outliers from the dolomites were from the same sample, drilled from the interior of a dolomite boulder. Four samples from a dolomite breccia produced a range of '813C








values that clustered in two areas, 21.6%o to 16.8%o PDB, and -21.2%o to -20.1%o PDB. Similarly, the oxygen isotopic data clustered at 3.8%o to 4.0%o PDB, and 6.7%o to 7.0%o PDB (respectively with the carbon clusters: not pictured).

The breccias contained a wide variety of isotopic compositions due to the mixed composition of the clasts (Figure 43). For cements, the broad range of carbon isotopic values fell between -34.4%o and -54.7%o PDB, with outliers at -22.5%o, -12.4%o, -1.2%o, and 15.0%o PDB. The oxygen isotopic values ranged between 2.9%oo and 6.6%o PDB, with no notable outliers. The breccia clasts did not cluster together, with a broad range of carbon and oxygen isotopic values occurring in many samples (-34.4%o to -45.0%o PDB and 3.0%o to 6.2%o PDB, respectively). One breccia contained all low Mg calcite clasts (determined through XRD), and the 813C and 180 for the clasts clustered around -35%o and 3%o PDB, respectively.
























Figure 5-1. Alvin photograph of the top of an aragonitic chemoherm from Hydrate Ridge. Field of view is approximately 8 meters.

























Figure 5-2. Alvin photograph of aragonitic crusts from Hydrate Ridge. Field of view is approximately 5 meters.
























Figure 5-3. Alvin photograph of high Mg calcite slabs from Hydrate Ridge. Field of view is approximately 4 meters.























Figure 5-4. Alvin photograph of Fe-Mn encrusted dolomite boulders (boulder field) from Hydrate Ridge. Field of view is approximately 10 meters.



















Figure 5-5. Hand sample (MS01) of chemoherm with multiple colors of dark grey and light tan carbonate and acicular aragonite. Upper box shows location of thin-section sample in Figure 19. Lower left box shows "banding" in acicular aragonite (see Figure 22).
























Figure 5-6. Hand sample (MSO3) of aragonitic crust with multiple colors of dark grey and tan cement. Lower box is location of thin section sample in Figure 26.





38












Figure 5-7. Hand sample (MSO5) of high Mg calcite slab with thin weathering rind. Upper box indicates the location of thin section shown in Figure 28.























Figure 5-8. Hand sample (MSO7) of dolomite boulder with Fe-Mn weathering rind.

























Figure 5-9. Hand sample (MS15) of breccia from Hydrate Ridge. Notice the sorting in parallel bands and the angularity of the clasts.










Sample: jj3 File: JJ3R31.SM ('


4-NOV-99 16:21


Corundum


5.00

4.05 3.20

2.45 1.80 1.25 0.80

0.45 0. 20


Corundum


Quartz


Corundum


I I


24.0 26.0 28.0 30.0


32.0


II I II


34.0


36.0


38.0


Figure 5-10. Example of XRD from drilled samples of hand samples. Notice the presence of only one carbonate mineralogy (high Mg calcite). The sample was mixed with a reference mineral (corundum). Secondary detrital minerals include quartz and feldspar. The position of the high Mg calcite peak between 290 and 310 20 was used to determine the mol % of MgCO3.


Mg Calcite


P I- I, I, L-1

































Figure 5-11. XRD analysis performed on core B20. Two centimeter intervals begin at the top of the graph and continue downward through the sediment column (0-2 cm is represented by the top line while 8-10 cm is represented by the bottom line). Large unlabeled peaks are corundum,.the internal standard used when running these samples. The stoichiometric dolomite peak falls at 31 degrees; most of these samples contain proto-dolomite, with minor amounts of stoichiometric dolomite. Notice all of these samples are comprised dominantly of dolomite and aragonite, with a Mg calcite constituent also present in every sample. The mixed mineralogy could be related to breccia clasts.








*% .',4* i a


2.ma"M


Figure 5-12. XRD results from core, CJ4. The top line represents the uppermost interval (0-2 cm), and each line below is a consecutive two centimeter interval. This core has dolomite at all intervals, with the addition of at least one extra mineralogy (aragonite, calcite, or Mg calcite). Mg calcite and aragonite tended to decrease with depth, probably indicating the replacement of those minerals with dolomite.
























Figure 5-13. Example of zoned clasts from core CJ4. These zones could be the reason for a variety of mineralogies present in one clast (as recorded by XRD).







G'a to
WS6,616


Figure 5-14. Example of zoned clasts from core B20. The clasts recovered from the cores were commonly breccias, as pictured in this sample.






46



100 Drk Aciulm o Grey a oniteHigh Mg Calcite Dolomi
nite Aragonite
80 70 c 60

0
e
350
C

20
a.
30 20 10

0
Sample #




Figure 5-15. Percent carbonate values (see Figures 7 and 8 for examples)
plotted by mineralogy. The aragonite is broken into dark grey, tan, and white
(i.e., acicular) colors, and they generally have the highest total percent
carbonate. The high Mg calcite samples (see Figure 9) span the broadest range of values, but the limited dolomite samples measured also had wide
variety. Every sample measured had at least 39% carbonate.









Table 5-1. Comparison of total mass percent carbonate with calculated volume percent carbonate. The calculated volume percent estimates the initial porosity present in the sediments. Notice that the aragonite samples have the highest initial porosity, and the dolomite samples have the widest variety in initial porosity.


Carbonate Type


Total Mass Percent Carbonate


Dark Grey Aragonite Dark Grey Aragonite Dark Grey Aragonite Dark Grey Aragonite Dark Grey Aragonite Dark Grey Aragonite Tan Aragonite Tan Aragonite Tan Aragonite White Aragonite White Aragonite High Mg Calcite High Mg Calcite High Mg Calcite High Mg Calcite High Mg Calcite High Mg Calcite High Mg Calcite High Mg Calcite High Mg Calcite High Mg Calcite High Mg Calcite High Mg Calcite High Mg Calcite High Mg Calcite High Mg Calcite High Mg Calcite High Mg Calcite High Mg Calcite High Mg Calcite High Mg Calcite Dolomite Dolomite

Cores
B2-4-1
B2-4-3
B2-4-4
B2-4-5
B8-10-1
B8-10-2
B8-10-3
B8-10-4
B8-10-5


57 79 86
79 76 77 69 78
67 91 70
72 63 69 94 88
73 54 39 83 69 70
84 79
84 '44
54 58 70 73 76
85 45


Volume Percent
Carbonate
55 77 85
77 75 75 67 76 65 90 68 70 60 67
94 87 71
52 37 82
67 68 82
78 83
42 51
* - 56
68 71
74 84 42


Single Clasts From Two Intervals









Table 5-1-continued Carbonate Type


CJ2-4-1 CJ2-4-2 CJ2-4-3 CJ2-4-4 CJ2-4-5 CJ8-10-1 CJ8-10-2 CJ8-10-3 CJ8-10-4 CJ8-10-5 B20-0-2 B20-2-4 B20-4-6 B20-6-8
B20-8-10 CJ4-0-2 CJ4-2-4 CJ4-4-6 CJ4-6-8 CJ4-8-10 CJ-10-12 CJ-12-14 CJ-14-16 CJ-16-18


Total Mass Percent Carbonate


Volume Percent Carbonate
55
0 3


Large Clasts From Two Centimeter Intervals









Volume %
40


0

2


Carbonate
60


UI




a 4 * Core B20 I a. Core CJ4
U



5U


Figure 5-16. Estimated carbonate volume percentage versus depth in core for two cores, measured from large, sieved clasts. The estimated porosity fluctuates with depth, and no clear patterns can be distinguished. Error bars are included on the two intervals with 6 data points.























Figure 5-17. Transmitted-light photomicrograph of large acicular aragonite fans radiating from micritic aragonite. This photo taken from a chemoherm sample (box in Figure 7). Notice the large amounts of pore space which enables the growth of long aragonite needles.























Figure 5-18. Aragonite clasts cemented by aragonite needles observed in chemoherm and crust samples (transmitted light photomicrograph). These clasts are angular and poorly sorted, pointing to in situ cementation.


























Figure 5-19. Radiating fan-like aragonite needles with blunt ends. SEM image taken from a chemoherm sample in the white aragonite zone. Notice the purity of the aragonite and the large pore spaces, supporting an origin above the seabed surface.























Figure 5-20. Transmitted light photomicrograph of a band in aragonite needles in box from Figure 7. These structures seem to be a result of dissolution and reprecipitation, creating darker bands where less aragonite is present.

























Figure 5-21. SEM image of a band in aragonite needles from chemoherm sample. The band formed during a time of dissolution and reprecipitation, which appears as a darker line along the needles (black arrow).




























Figure 5-22. Small aragonite crystals growing from a micritic aragonite in a transmitted light photomicrograph of a crust. The change is fairly abrupt between the elongated needles and the smaller micritic aragonite, possibly caused by a change in water chemistry and carbonate supersaturation. Micritic carbonate was determined to be aragonite and not high-Mg calcite based on EDS sampling that revealed Sr and no Mg.


























Figure 5-23. Boundary of aragonite matrix and clamshell (lower right) in an SEM photomicrograph. The aragonite is primarily micritic, but some radiating fans are present. The boundary is poorly defined, and the aragonite is consistent in texture both proximal and distal from the clamshell.























Figure 5-24. Thin section photomicrograph of light tan and dark grey micritic aragonite commonly seen in chemoherm and crust samples (see Figure 8 for location). Whole and fragmented forams are commonly included in all the


























Figure 5-25. SEM image of detrital material is common in the tan and grey aragonite cement. Clays and quartz are the most common detrital material included in the carbonates.























Figure 5-26. Typical cement of a high Mg calcite sample seen in a transmitted light photomicrograph with broken forams, glauconite nodules, and low porosity.


























Figure 5-27. Example of micritic high Mg calcite in SEM photomicrograph. The crystals are all extremely small and poorly formed. This evidence supports an origin in an area with limited amount of room to grow (the sediment column).

























Figure 5-28. Detrital material cemented by high Mg calcite seen in SEM. This quartz grain is cemented by poorly formed high Mg calcite.


























Figure 5-29. This high Mg calcite sample (MS 13) had the highest percentage of detrital material in any sample analyzed by SEM. Diatom frustules and clays are the most common detrital material in this image.
























Figure 5-30. Banded micritic high Mg calcite and small clasts of pyrite, glauconite, and detrital material. The small clast zones include glauconite, and pyrite is present in both the micrite and the clast zones. This structure could be related to bands in the sediment during precipitation.
























Figure 5-31. Dolomite sample with multiple veins. The micritic dolomite appears to be fractured and then reprecipitation occurs in the open spaces. Notice the very low percentage of pore space present in the thin section.




























Figure 5-32. Dolomite crystals highly magnified under SEM. The crystals are similar shapes and size, with some detrital material present.


























Figure 5-33. SEM image of the boundary between a dolomite clast (upper left) and a dolomite vein (lower right). The two matrices are indistinguishable, implying a similar precipitation environment for both. Notice the small percentage of pore space in the cement.
























Figure 5-34. Transmitted-light photomicrograph of a representative breccia commonly found on Hydrate Ridge. The clasts are subangular, and comprised of high Mg calcite. The cement is acicular aragonite. High amounts of pore space allow for acicular aragonite growth and suggest an origin at or near the seabed surface.
























Figure 5-35. Breccia with large subrounded clasts, glauconite nodules, and shell fragments seen in transmitted-light photomicrograph. Aragonite needles cement the clasts and shells. Notice the large percentage of pore space.
























Figure 5-36. Transmitted-light photomicrograph of breccia with large clamshell, clasts, glauconite, and forams. Further support for an origin of breccias at the seabed surface is the large whole forams present in thin section. These also support an in situ formation as opposed to a mass wasting event because the forams are unbroken.
























Figure 5-37. Pyrite infilling pore space in a high Mg calcite sample (transmitted-light photomicrograph). These samples are much less porous and include more pyrite than the aragonitic samples.
























Figure 5-38. High Mg calcite sample with different colored vein seen in transmitted-light photomicrograph. The different colors are generally the same material with common forams and pyrite. Changes in color could be caused by differences in the amount of detrital material. Notice the low amount of pore space in the micritic cement, supporting precipitation in the sediment column.
























Figure 5-39. Edge of high Mg calcite sample (located at the top of the photo) with large glauconite nodules as seen in transmitted-light photomicrograph. Dissolution of the glauconite from water flowing through the carbonate creates the brown hematite staining seen in this slide.


























Figure 5-40. SEM image of framboidal pyrite (as identified by SEM EDS) seen in a dolomite from Hydrate Ridge. Pyrite is common in dolomite and high Mg calcite samples, as seen in thin sections.





















-50
*


C.*
*
-45
I X X
* C
*



35
-40






-35


2 25 3 3.5 4 4.5 d180 (ppt PDB)


* Acicular Aragonite gDark Grey Matrix a Chemoherm Dmark Grey Matrix x Light Tan
*Tan Grey Mix e Breccia cements


5 5.5 6 6.5


Figure 5-41. Stable isotopic values for aragonitic samples and breccias. Acicular aragonite has less positive 8180 values than other cements. Dark grey cement trends from more negative 813C and lower 8180 to less negative 813C and higher 6180. Chemoherm dark grey cement is relatively negative compared to other samples. The light grey and mixed cements have the highest 8180 values. The breccia cements are extremely random, with no obvious trends.






























Figure 5-42. Authigenic aragonite and clamshell isotopic compositions. The pure acicular aragonite has more depleted 81O80 values than the darker grey aragonite, implying a greater influence of seawater during the pure aragonitic precipitation. Notice that the grey and pure aragonite samples trend toward a common value (near 4 ppt PDB) with more negative 813C values. The less negative 813C values for the pure and grey aragonite show a wide separation between 180O values for each. Clamshells were plotted as a point of reference and as an example of typical isotopic signatures of seawater at 4oC.































Figure 5-43. Isotopic signatures for high Mg calcite samples. The 8180O values are all positive, reflecting a complicated signature possibly associated with high temperature fluids during precipitation. The 813C reflects the influence of methane-rich waters during precipitation. The 813C has a relatively wide range of values, which probably reflects a mixing of diagenetic marine water with biogenic methane waters. The isotopic values fall into three broad groups which reflect slightly different influences during precipitation.














3419 CS5-MS05
Transect l


Transect 2 del C-13 del O-18 PDB PDB
MDSO86 Exterior -44.8 4.46 MDSO87 -43.72 4.42 MDSO88 -40.19 4.34 MDSO89 -41.89 4.43 MDSO90 -34.93 3.66 MDSO91 Interior -39.94 4.10

Figure 5-44. Shows table of high Mg calcite samples from cross sections for isotopic compositions. These samples were drilled across two calcite slabs (only one slab is shown). Transect 1 cross section depicts a clear trend in 813C values that decrease to the base of the slab. The 5180 values for the cross section are generally consistent, with a slight positive trend toward the base of the slab. Transect 2 813C values generally become less negative toward the base of the slab. The 6180 values decrease slightly toward the base of the slab.





















-15




-20




-25


i



a*
S.-30




-35 -40




-45
3 5 7 9 11 13 15 17 19 d1O80 (ppt PDB) Figure 5-45. Isotopic signatures for dolomite samples. The 813C is less

negative than values from aragonite or high Mg calcite samples. All 8180

values are positive, with the majority of the points falling between 5 and 7 ppt

PDB, and one outlier at -41%o PDB (not shown).













CHAPTER 6
DISCUSSION

Published Models of Carbonate Precipitation

One general model has been suggested for the precipitation of authigenic carbonates by several authors. Beauchamp et al. (1989) proposed that North Sea authigenic carbonates formed in the sulfate reduction zone. They also presented an in situ origin for the brecciation of carbonate crusts caused by high gas pressure (also causing the pockmarks seen in association with active venting). The pyrite found in most samples was attributed to occasional hydrogen sulfide seepage from deep in the sediment column (Beauchamp 1989). All three mineralogies (aragonite, high Mg calcite, and dolomite) present in carbonates from the Gulf of Mexico were studied by Roberts and Aharon (1994). They offered a model of formation for all precipitates through bacteriallymediated reactions in the sulfate reduction zone or at the seabed surface (Roberts and Aharon 1994). Authigenic carbonates from the Florida escarpment are composed primarily of high Mg calcite, but aragonite and dolomite are present. Paull et al. (1992) suggested a general theory of formation similar to Roberts and Aharon (1994) where oxidation of methane and resulting sulfate reduction and increasing alkalinity promoted carbonate precipitation (Paull et al. 1994). These general models mentioned above provide an explanation for carbonate precipitation; however, they do not offer explanations for the presence of multiple mineralogies as seen on the Cascadia margin.

Greinert et al. (2000) focused specifically on types and origin of the carbonates on Hydrate Ridge, and proposed precipitation in the near surface, slightly reducing conditions







in the top few centimeters or on the seabed, or from in situ brecciation of the mudclast and intraformational brecchias through slumping or venting. Aragonite precipitation requires near sediment-water interface conditions where high sulfate concentrations are present, while high Mg calcite precipitates where sulfate is reduced (sulfate reduction zone, >3 cm depth). Calcite is the favored phase at temperatures found on Hydrate Ridge, but the presence of sulfate greatly inhibits calcite precipitation and favors aragonite (see Literature Survey section). Greinert et al. (2000) suggested that the dolomite mudstones originate at the deepest depths, in the lower methanogenic zone and below (10-220 mbsf). They suggested this depth based on oxygen isotopic values that are highly negative (-3.5%o SMOW), and likely to occur at deeper depths where temperatures are higher (minimum 15.3oC).

Kulm et al. (1986) proposed a general origin for all carbonate cements at a shallow to surficial (upper 3 meters) depth based on 8180 values. Positive 8 80 values indicate low temperature waters, which would occur at shallow depths (Kulm et al. 1986). Bohrmann et al. (1998) suggested that aragonite collected from chemoherms reflect recent seabed surface conditions, based on positive 8180 values (3.68%o PDB). However, they indicated that the high Mg calcite precipitated as a result of gas hydrate destabilization that occurred some time in the past, possibly caused by sea level lowering or an increase in temperature of the ocean. This conclusion was based on 180 values which indicated high Mg calcite precipitated from porewater equilibrated with isotopically heavier (+3%o, from Bohrmann et al. 1998) water derived from the release of cage waters during dissociation. The oxygen isotopic values for high Mg calcite in their study were 4.86%o PDB which is enriched compared to average seawater (+.95%o SMOW). Borhmann et al.







(1998) suggested that the 8180 enrichment was caused by the influence of hydrate-rich pore fluids (+3.0%o SMOW).

Previous investigators have promoted a variety of explanations for the different

types of carbonates found on the Cascadia margin. Most papers agree that the carbonates are forming in the upper few tens of meters from the seafloor surface. However, they disagree on precisely where, and when the precipitation occurs. Morse et al. (1997) and Bohrmann et al. (1998) proposed a fluctuating system of high Mg calcite precipitation or aragonite precipitation based on climate and gas hydrate changes (Bohrmann et al. 1998, Morse et al. 1997). This explanation could only be applied to precipitation at the sediment surface because aragonite is unlikely to be the favored precipitate in the sulfate reduction zone due to the low temperatures and depleted sulfate (Burton and Walter 1987). A more likely explanation is high Mg calcite precipitating at depth while aragonite precipitates at the surface. The few authors that dealt with breccia formation on the Cascadia margin support a similar origin as presented herein, which suggests that gas hydrate destabilization causes either in situ brecciation or a massive gas discharge in the shallow seabed (Beauchamp et al. 1989, Greinert et al. 2000). This theory is based on the presence of pockmarks, observations of floating gas hydrates, and analysis of the clasts and cements of the breccias (Bohrmann et al. 1998). The dolomite is often believed to precipitate at greater depths, up to 40 mbsf (Greinert et al. 2000). If precipitation occurs at this great of depth, exposure of the many large boulders on the seafloor surface of Hydrate Ridge would be difficult to explain through currents or gas hydrate destabilization which only involves the upper few meters of the seabed.








Conceptual Model of Carbonate Precipitation

The processes leading to the precipitation of carbonate phases on Hydrate Ridge are complicated, and involve multiple influencing factors including sulfate concentrations, temperatures, methane concentrations, and carbonate saturation state. The primary controls on formation of aragonite versus high Mg calcite promoted in this study are temperature of water during precipitation, porewater sulfate concentrations, and carbonate saturation state. The sulfate conditions are particularly important, because sulfate-rich or depleted zones can indicate depth of precipitation (Baker and Kastner 1981, Burton and Walter 1987). Aragonite can only precipitate at the temperatures measured on Hydrate Ridge (4oC) if the precipitation occurs in the sulfate-rich seawaters, or if precipitation occurs in conjunction with warm-temperature migrating fluids seeping up from depth (Burton and Walter 1987). Even slightly higher than average temperatures (+0.320C) have only been recorded in a few places on the Cascadia margin because the high temperatures of migrating fluids are hard to maintain (Kulm et al. 1986). The warm waters quickly equilibrate with the surrounding colder porewaters through diffusion in a matter of days. Aragonite is usually found at the sediment water interface or in the water column as a chemoherm, and therefore the influence of sulfate is probably greater than migrating warm waters. The zone of sulfate reduction generally occurs below the upper few centimeters of the seabed, so the aragonite precipitation is not favored in this region (Greinert et al. 2000), and high Mg calcite, which is the favored phase at these temperatures, should precipitate in this zone (Figure 6-1).

Dolomite precipitation is more difficult to understand where and how it occurs, or if the dolomite is authigenic (direct precipitation from seawater) or diagenetic (alteration







from calcite). Because sulfate greatly inhibits the dolomitization process, precipitation must occur somewhere in the sulfate reduction zone in the upper tens of meters from the seabed surface or below that where no sulfate is left. Baker and Kastner (1981) experimentally found that even minor amounts (0.001 M) of sulfate greatly reduced the precipitation of dolomite. Minor sulfate has been found at depths as great at 60 mbsl on Hydrate Ridge (Westbrook et al. 1994). Baker and Burns (1985) also promoted a shallow sediment depth for dolomite precipitation, based on high magnesium and low sulfate porewater concentrations. Overall, the literature generally suggests that dolomitization can occur through replacement or authigenic precipitation, which can take place at depths up to a few tens of meters (Baker and Burns 1985, Baker and Kastner 1981).



Aragonite Phase

The thin section and SEM work are particularly important in understanding where mineralogies form. Aragonitic chemoherm samples have the largest amount of voids and the longest needles, which indicates high amounts of pore space and therefore precipitation at or above the seabed. They also have the largest percentage of carbonate (indicating the highest porosity), and a significant portion of pure aragonite (determined through XRD analysis). The presence of pure aragonite and absence of detrital material can only be achieved at or above the seabed surface. The different colored carbonates seen in chemoherm samples were composed of the same microcrystalline aragonitic and detrital mixture. This is important in establishing that no significant formational changes occurred to produce a mineralogy other than aragonite. Instead, the darker cement could indicate a higher percentage of detrital material, or reduced porosity due to dissolution and







reprecipitation of microcrystalline aragonite (the percent carbonate analyses in this study were inconclusive). Micritic aragonite could form by recrystallization even if precipitation occurs at or above the seabed where space is available for large crystals to grow. Reid and Maclntyre (1998) suggested that micritic aragonite could precipitate in pore spaces or by recrystallization. The aragonitic needles seen in SEM and thin section seem to have banding, but on closer inspection the "bands" are gaps in the needles, probably indicative of periods of dissolution and then regrowth. SEM images also reveal the relative amounts of detrital material in the aragonite (less detrital) versus other phases. The low-detrital aragonite suggests precipitation at the seabed surface or above it. The highest percentage of whole forams is present in the aragonite samples, which indicates very little post-depositional compaction. All of these factors support a formation at or above the sediment water interface. One line of evidence that could support precipitation of aragonite at the sediment water interface is the absence of hematite staining and glauconite. If aragonite is forming at or slightly below the seafloor surface, then these samples should have the highest percentage of hematite staining due to the diffusion of seawater. Instead, the aragonite samples have almost no hematite stains present. This could imply a formation within a highly oxygenated zone, but recent precipitation has not allowed the oxidation of the pyrite and glauconite.

The carbon isotopic data for the aragonite samples shows the strong influence of methane-rich waters during precipitation, implying that precipitation occurs near vents without too much mixing with seawater. The oxygen isotopic values of the aragonitic samples suggest the influence of seawater, with one sample averaging at a low

4.06%o PDB while the highest averaging at 4.89%o PDB. The IO8 values of the








formation water are calculated values of the 8180 during precipitation using standard equations discussed in the Methods section. Formation water 8180 values from the chemoherm samples and the aragonitic samples ranges between -1.1 and 0%o PDB and

-0.3 and 0.8%o PDB respectively (recording the lowest 8180 formation water values), implying precipitation influenced primarily by seawater (Grossman and Ku 1981).

The modes of aragonite precipitation presented here are similar to the observations of others. Ritger et al. (1987) proposed a theory of aragonitic needles forming after the micritic calcite cement, which is only observed in this study amongst breccias. The micritic aragonite implies either recrystallization occurred or during precipitation there was little room for growth. The aragonite needles imply large open pore spaces and an origin above the seabed. The distinct boundary between the cement and needles could also point to a change in conditions during formation, or the inclusion of different amounts of detrital material (Ritger et al. 1987, this study). Bohrmann et al. (1998) and Greinert et al. (2000) proposed that the aragonitic needles grow into gas hydrate cavities. This theory could not be tested in this study, but is supported by the highly depleted carbon isotopic signature of the aragonite and the large cavities seen in thin section. Kulm et al. (1986) found well-sorted glauconite associated with the aragonite cement. However, the glauconite in the samples studied here is poorly sorted, and sometimes infrequent, and could indicate recent precipitation.

This study also supports Ritger et al.'s (1987) correlation between higher

induration and higher total carbonate percentage, which includes all three carbonate mineralogies. Ritger et al. (1987) noted two groups of samples, one with high mass percent CaCO3 (25-30%) and the other with minimal carbonate (less than 20% CaCO3).








This bimodal pattern in Ritger's samples does not occur in the samples in this study, with the possible exception of clasts from the cores. Several clasts from cores contain practically no carbonate, while the remainders have a high CaCO3 percentage (Figure 5-45). Ritger's association between high induration and high carbonate content is also recorded in this study, and is important in determining where the carbonate precipitated, as high detrital content reflects precipitation in the sediment column. Sample and Kopf (1995) reported much lower total carbonate percentages than Ritger or this study, with the highest values at 25% CaCO3.

Kulm et al. (1986) noted a difference in carbon and oxygen isotopic values of

aragonitic cements versus aragonitic crusts. These trends are not seen in the data from this study, and could indicate a similar formation process for both the cements and the crusts (Kulm et al. 1986). Greinert et al. (2000) also recorded positive carbon isotopic values for chemoherm samples, where the chemoherm samples from this study center around

-45%o PDB, with no outliers. Greinert et al.'s positive carbon isotopic values could reflect the influence of seawater or higher temperatures on chemoherm samples, while this study's negative 83C values imply some influence of methane-rich waters.



High Mg Calcite Phase

High Mg calcite is favored for precipitation at 4-10 cm depth in the zone of sulfate reduction. This depth of precipitation is based on the calculated porosity at the time of precipitation (<10 cm) and the location of the top of the sulfate reduction zone'(>4 cm). High Mg calcite is considered to be the dominant carbonate mineralogy present on the Cascadia margin, but some carbonates could have a high enough mol%Mg (>30








mol%Mg) to be considered proto-dolomite. As stated previously, the Greinert classification system will be used in this paper since it is commonly used with carbonates from the Cascadia margin (high Mg calcite- 8-20%; proto dolomite- 30-40%). Ritger et al. (1987) had a broader range of values for high Mg calcite (6-20 mol%Mg) than this study (6-11 mol%Mg), which could indicate a continual range of mol%Mg from high Mg calcite to dolomite. Sample and Reid's (1998) findings supported the minor constituents identified in this study in high-Mg calcites, which include clays, feldspars, and quartz. They identified the dominant cements as high Mg calcite and dolomite, partially concurring with this study. The same authors concluded that high Mg calcite was the primary cement, with a mol%Mg of 3 to 12. However, these conclusions were based on a small population size (35 samples) that was not sampled in a statistically random fashion (Sample and Reid 1998).

This zone of precipitation (4-10 cm depth) is supported by reduced carbonate mass percentages (implying less porosity), higher detrital material seen in thin section and SEM, and much less void space than aragonitic samples. Sediment porosities determined by gravimetric analyses of Alvin push core subsamples are 0.5-0.7 (50%-70% carbonate by mass) in the range of 4-10 cm below the seafloor (Kastner pers. comm.). The presence of micritic calcite implies little room for crystal growth, and is cited as a support for high Mg calcite precipitation in the upper sediment column where little pore space inhibits crystal growth (Kopf et al. 1995). A much higher percentage of high Mg calcite thin sections include broken forams than seen in aragonitic samples. The lack of hematite stains and glauconite in the high-Mg calcite samples supports a shallow burial, where water advects less than at the seafloor where water moves freely. Carbon isotopic values








also show the importance of methane-rich waters during precipitation, with a range from

-52%o PDB to -30%o PDB. The i180O values from high Mg calcite samples fall between

5.18%o PDB and 4.13%o PDB. These very positive oxygen isotopic values reflect the importance of hydrate-rich waters during precipitation, and imply precipitation at some depth in the sediment column where seawater is not as influential. For example, the water in which the high Mg calcite formed has a calculated 180O value of +0.7 to 2.7%o PDB at 3.90C (Tarutani et al. 1969). When methane hydrates dissociate, they release cage waters with 8 l0 of 2.7%o SMOW and therefore have a significant influence over the 180O values of the high Mg calcites.

The high Mg calcite and dolomite observed in this study are similar to those described by given by Ritger et al. (1987). The cement is primarily microcrystalline calcite and dolomite, with less pore space. Sample and Reid (1998) found that the calcite cement often included unbroken forams, bivalves, and veins, all of which are seen in this study. The veins have detrital grains accumulating in them, implying precipitation in the sediment column where grains can move into the veins. Most authors studying Cascadia carbonates did not report any hematite staining in their thin sections, and dolomite descriptions and formation were rarely discussed. Our study found multiple hematite stains in a variety of carbonate types (high Mg calcite and dolomite). This could indicate fluid flow and the presence of oxygen, and therefore whether the hematite formed at depth or at the seabed surface. Unfortunately, hematite staining could occur during formation or after exposure to the shallow sediment column, and cannot conclusively indicate depth of formation.








Ritger et al.'s (1987) group of high-Mg calcite samples all reflected a methane influence (highly depleted ~'13C values). Their values are similar to the ones from this study, except the range is larger and includes much lower values (as depleted as

-66.7%o PDB, while this study's lowest sample is -53.2%o PDB). The Ritger et al. study also recorded relatively high oxygen isotopic values, similar to this study, with a range between +3.66 and +7.22%o PDB. The high oxygen isotopic values could be due to temperature changes or multiple events of hydrate formation and dissociation. Kulm and Suess (1990) also observed a broad range of negative 813C values in high Mg calcites (down to -66.7%o PDB). The positive oxygen isotopic values seen in Ritger et al., Kulm and Suess, and this study are probably indicative of a series of hydrate formation and dissociation, and dolomitization (Kastner, pers. comm., Kulm and Suess 1990, Sample and Reid 1998).



Dolomite Phase

Dolomitic samples are common in this study, and are the dominant carbonate type in the cores. This result is surprising because authigenic dolomite precipitation is difficult to explain because experimental studies indicate that precipitation occurs in small amounts (Baker and Kastner 1981), and is not the dominant carbonate in Alvin-collected hand samples. Malone et al. (1996) suggested that authigenic dolomite in the Monterey Formation of California precipitated at times when migrating fluids in the shallow sediments reached temperatures over 1000C. This theory could be applied to the Cascadia margin if periodic fluxes of warm fluids migrated to the seabed in the past 1-24 ky (dates from Kastner et al. 2000). The possibility of large amounts of authigenic dolomite








precipitating under modem conditions is difficult to explain because of colder temperatures; however, replacement dolomite could be a viable explanation of the large amounts of dolomite seen on Hydrate Ridge because high Mg calcite and aragonite will be replaced by dolomite. Unfortunately, replacement dolomite is impossible to distinguish from authigenic dolomite, and cannot be confirmed in this study. Because sulfate inhibits dolomitization, dolomite must form in the zone of sulfate reduction (where pockets of complete sulfate depletion are present) or below it (where sulfate is not present), in the upper tens of meters of the sediment column. Microenvironments of sulfate concentrations near zero can occur in the upper meters, and could be zones of dolomite precipitation.

This study proposes a dolomite origin within the upper couple of meters of the

sediment column, possibly at the same level as the high Mg calcite samples (>3 cm depth). This relatively shallow depth of formation is based on the amount of dolomite seen on the seabed surface and the availability of magnesium and calcium which is greatest in the upper couple of meters. Dolomites from Hydrate Ridge have an average of 61 mass percent CaCO3, which indicates less porosity than aragonite or high Mg calcite samples. The low porosity, fragmented forams, and micritic cement support precipitation in the sediment column (1-10 cm depth). The presence of hematite staining and iron-manganese coating on many of the dolomite samples could attest to the old age of the dolomite and the movement of water through the dolomites since exposure to oxic seawater for a long period of time creates these rinds. Studies on iron-manganese crusts on the continental shelf off Tasmania suggest that these grow at rates of approximately 10m/200,000 years (Exon 1997). The growth rates must be greater on the Cascadia margin since active








venting has only occurred for the past 21-24 ky, but still point to an old age for the dolomite boulders (dates from Kastner et al. 2000). The carbon isotope values from dolomitic samples range between -34.1%o and -39.6%o PDB, which reflect some influence of methane-rich waters. The 818O values for the dolomite fall within 5.5%o and 7.2%o PDB. Formation water 8180O values of dolomite samples fall between 1.1 and 2.8%o PDB, respectively.

Sample and Reid (1998) reported isotopically similar dolomite samples to the ones in this study, as well as a group with slightly negative carbon isotope values (-1%o to

-25%o) and negative oxygen isotope values (-4%o to -13%o). They attribute the differences in the isotopic values to changes in migrating fluids in different areas. This explanation could result in different samples with unique isotopic signatures dependent upon where the samples precipitated, which could explain some of the outlying carbon and oxygen isotopic values in their data and data presented in this study. Unfortunately, no systematic differences in the outliers exist and cannot directly support different locations of precipitation.

Some aberrations exist within the data that does not fit with the model presented in this paper. These include no hematite staining in aragonite samples, and common hematite staining in dolomite samples. Also there are no clear trends in 813C volume percent carbonate values to suggest great influences of methane with depth. Finally, the presence of dolomite in all intervals of cores also does not fit the model of dolomite precipitating at depth as presented above.








Breccias

Authigenic carbonate breccias have been separated into a wide variety of

categories by previous authors, including mudclast breccias, intraformational breccias, sandstones, and mudstones (Greinert et al. 2000, Kopf et al. 1995). The breccias collected for this study primarily fall into mudclast and intraformational breccias. Intraformational breccias are the most common in this study, as many contain multiple types of clasts and shells, and are cemented by aragonite. The majority of the breccias analyzed are composed of detrital and high Mg calcite clasts with acicular aragonite cement.

The breccias could be related to destabilized gas hydrates originating in the sediment column. The primary factors that support this conclusion come from observations in thin section and hand specimens. The breccias are composed of subangular to subrounded clasts, which are very poorly indurated. These observations are important because any transportation of these clasts would have rounded them or destroyed them completely. The matrix also commonly includes whole forams, which would have been crushed with any mass movement, such as a slump or turbidite flow. The high percentage of voids in breccia samples and the dominance of aragonitic cement in the breccias suggest cementation at or near the sediment water interface. Destabilized gas hydrates could break apart weakly cemented clasts, possibly sending material up into the water column and creating the pockmarks seen on Hydrate Ridge. The breccias seen on Hydrate Ridge could also have formed in situ through faulting or fracturing, which requires no transportation of clasts. Another possibility suggested by Suess et al. (1999) is that floating hydrates can carry sediments and clasts with them and then release them as the hydrate dissociates. This material settles down to the seafloor, where it is eventually




Full Text

PAGE 1

METHANE-RELATED AUTHIGENIC CARBONATE FORMATION ON THE CASCADIA ACCRETIONARY PRISM By MARY LINDSEY BATEMAN A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2002

PAGE 2

. LD 17 8 0 20 D2 J <.,./ D

PAGE 3

Copyright2002 by Mary Lindsey Bateman

PAGE 4

For my parents and grandparents , whose support and love made this thesis possible

PAGE 5

ACKNOWLEDGMENTS I first want to thank Dr. John Jaeger for his constant input and help with the research and writing of this thesis. The other members of my committee, Drs. John Martin, Anthony Randazzo, and Kenneth Sassaman also have been invaluable for referrals and discussions. I would like to extend my appreciation to Kendall Fountain and iason Curtis for their help running samples and for their useful suggestions. I would also like to thank my parents and my grandparents, who have been my strongest support during my years at UF. Their guidance, patience, and unconditional love ultimately were the driving forces behind this thesis. Aleta MitchellTapping and Timothy Gysan also played an integral part in supporting me and helping with proofreading and graphing. Thanks also to Marc Shook, Christy Gysan, Elena Miranda, and Mark Leidig, who were a valuable outlet for discussions during research and writing. Finally, I would like to thank all my close friends who have always given me support and love, and without whom my career would not be the same . . IV I

PAGE 6

TABLE OF C ONTENTS A CKNOWLEDGMENTS ... ..... ... ................ ... ... ........ ... . ............................................. ... ..... i v .......................... . ....... . .... . ................ . .................................. .............. . ... . ... ...... vi CHAPTER 1 ........................................................ . ............... ...... .... .... ... ..... ...... ... ..... 1 2 SruDY AREA (SITE LOCATION ) .......... . ..... . . . ...... . ... ... ................................... ........... 7 3 Tl_TRE Sl_TRVEY (BACKGROl_TND ...... .......... . ..... ..... ....... 9 .................................................................................... . ............................ .... .............................. . ...... . ....... . ....... . ...... . ...... . ..................... . ....... . ..................... () ........................... .......... .. ....................... ... ... . ... ........... ........ ........... ... .... ... ......... . .... . .............. .......... ...... ..... .. ......... ... ...... . ................ ... ... . ... ........... APPENDIX DATA FROM ....... ... ... . ............... ... ........ .................................. 97 <:>14 ......................... . ........................ . ....... . ...... . ..... . ... . . . ... . ... ..... ...... BIOGRAPHICAL SKETCH . .............. . .............. . ........................................................... v

PAGE 7

Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science METHANE-RELATED AUTHIGENIC CARBONATE FORMATION ON THE CASCADIA ACCRETIONARY PRISM By Mary Lindsey Bateman December 2002 Chairman: John Jaeger Major Department: Geological Sciences Methane-related authigenic carbonate formation has been documented in the rock record as well as in modem oceans around the world . Knowledge about the unique conditions under which the carbonate forms is important to understanding fluid flow on continental margins and the fate of carbon. The examination of the processes influencing carbonate on Hydrate Ridge is the basis for a Masters thesis in geology. Hydrate Ridge, an elongated ndge on the accretionary prism located off the coast of \ Oregon, represents a SJ;Uall part of the subduction zone between the Juan de Fuca and North American plates. High organic carbon fluxes result in widespread methanogenesis in the prism sediments producing large stores of methane hydrates . Thrust faults cutting the accretionary prism serve as conduits for the methane to migrate upwards to the top of the seabed, where it is trapped as gas hydrates or is released into the water column. Three carbonate mineralogies are present on Hydrate Ridge : aragonite, high Mg calcite, and dolomite . This thesis presents a hypothesis on the depth of origin of the . Vl

PAGE 8

.. Vll different mineralogies based on a variety of influencing factors . Analyses performed on the samples to understand their origin included optical and scanning electron microscopy, X-ray diffraction , carbon and oxygen isotopic analyses , and determination of the total mass percent carbonate . These analyses were aimed at understanding the sedimentary conditions under which the carbonate precipitated, and included porosity, availability of oxygen and sulfate, and methane concentrations as either free gas or hydrates. The primary factor controlling the carbonate mineralogy promoted in this study is the amount of dissolved sulfate in the formation waters. Aragonite precipitates at the seabed surface where sulfate concentrations are high . High Mg calcite preferentially precipitates where sulfate is depleted, in the upper sediment column at 4-10 em depth. Dolomite precipitation is also inhibited by sulfate, and therefore its precipitation occurs in the sulfate-reduction zone in the upper couple of meters of the seabed, as shallow as 4 em depth . The high concentration of dolomite present on Hydrate Ridge could be a result of recrystallization from high Mg calcite and precipitation in larger quantities in the past due to higher temperatures of migrating fluids (over 1 00C), or a result of pockets of extremely depleted sulfate in the sediment column . Breccias are common on Hydrate Ridge, and can be composed of high Mg calcite and aragonite clasts . They are proposed to be the result of hydrate destabilization , which fractured carbonates that then were re-cemented . The presence of pockmarks on Hydrate Ridge is evidence of this destabilization .

PAGE 9

CHAPTER 1 INTRODUCTION Authigenic carbonates play an important role in our understanding of overall fluid flow through convergent continental margins . Tectonic convergence on these margins leads to dewatering of sediments accompanied by active fluid flow along fractures and faults . This fluid flow has received much attention for its contribution to global geochemical fluxes, especially the global carbon cycle (Aharon 1994 , Carson et al. 1994, Kastner et al. 2000). Methane is commonly produced on convergent margins by the burial of organic material from both biogenic and thermogenic methanogenisis. Methane within continental margin sediments can be found in three phases according to local temperature and pressure. At sufficiently high temperatures (9 . 5C) and pressures deep in the sediments (20-2000 mbsf), methane is found as a free gas (Hyndman et al . 1995, Kvenvolden 1995) . As pressures and temperatures drop with a rise toward the seafloor, methane gas binds with porewaters to form methane hydrates . The transition from free gas to gas hydrate deep in the sediments approximately coincides with the bottom simulating reflector (BSR) . Methane gas can also migrate upward through faults and fractures and can mix with the seawater in the water column. Methane-oxidizing bacteria use free methane as an energy source, creating bicarbonate or C02 as a byproduct of the methane oxidation reaction depending on the oxidant used (sulfate or oxygen, respectively; specific equations discussed in Chapter 3) . The pore water in the shallow sediment column and the near-bottom seawater becomes saturated 1

PAGE 10

2 with bicarbonate , creating a favorable environment for the precipit a tion of carbonates (Ritger et al . 1987 , Suess et al. 1998). Methane seeps and related authigenic carbonates occur around the world (in both deep and shallow water) including the Gulf of Mexico and North Sea, and off the coasts of Denmark, Florida, Cascadia, and Barbados (Jorgensen 1992, Kastner et al . 2000, Martin et al . 1996, Paull et al . 1992, Rad et al . 1996, Suess et al . 1998 , ) . These methane seeps often occur in accretionary prisms associated with plate subdu c tion, and therefore the presence of authigenic, methane-related carbonates may prove useful in identifying similar deposits in the rock record. Ancient suspected cold-seep carbonates have been recognized in the rock record across the world (Canada , Washington State, and the Apennines) (Beauchamp and Savard 1992, Cambell 1992 , Terzi et al . 1994 ). Globally, methane related authigenic carbonates are composed of surprisingly similar mineralogies that include aragonite, high Mg calcite, and dolomite, which all appear regardless of hydrate occurrence or water depth. This similarity in mineralogies suggests a simple, common set of ubiquitous processes control the precipitat i on of different mineralogies . To better define the conditions in which methane-related carbonates form, a systematic understanding of fluid flow and corresponding methane dynamics in prism sediments is necessary. One region where a basic understanding of these processes exists is on the Cascadia Margin . In a field site known as Hydrate Ridge (Figure 1-1) (Bohrmann et al . 1998), a series of studies were conducted to evaluate fluid flow along faults (Westbrook et al. 1994) . Early work in this region by Kulm et al . (1986) recognized a connection between active methane seeps at the seafloor and the widespread occurrence of authigenic carbonates (Carson et al. 1994) . Recent work on Hydrate Ridge

PAGE 11

3 suggested a close relationship between gas hydrates and authigenic carbonate formation. The carbonates are concentrated where methane-saturated fluids seep from the seabed, and are hypothesized to form by the aforementioned reaction between methane and sulfate, mediated by microbial communities . Behrmann et al. (1998) recovered carbonates within gas hydrates, and proposed carbonate precipitation in and around the hydrates in the upper 50 em of the sediment surface. Yet, this connection cannot fully explain the global similarities in carbonate mineralogies (aragonite, high Mg calcite, and dolomite) and morphologies (chemoherms-discussed in Study Area section, crusts, slabs, boulders, and chimneys), especially in regions where gas hydrates are not present. Because of the global similarity of carbonate mineralogies and morphologies on continental margins, the processes leading to precipitation of the three mineralogies must be equally common. Simply, mineralogy is influenced by the kinetics of precipitation, whereby a particular phase is favored. Under most porewater conditions at ocean bottom temperatures, Mg calcite is the favored form; but when even minor amounts of sulfate are present (greater than 5% of seawater value), aragonite is favored (Baker and Kastner 1981, Burton and Walter 1987) (Figure 1-2). Calcite is also inhibited by high concentrations of hydrated Mg2+ ions (present in ocean water), promoting aragonite precipitation instead of calcite (Ritger et al. 1987). In this setting on Hydrate Ridge, sulfate is most common in the seawater and shallow sediments, but is rapidly consumed at depth in the sediment column by methane oxidation. This system is likely common in methane-rich margin sediments . The goal of this thesis is to investigate the formation of authigenic carbonates related to methane seepage and possibly related to hydrate formation and dissassociation

PAGE 12

4 on Hydrate Ridge by using carbonate samples and cores collected from the seabed by DSRV Alvin. More specifically , the hypothesis presented in this thesis is that aragonite is the favored form precipitated at or near the seabed where sulfate-rich waters are present. Other phases (high Mg calcite and dolomite) form deeper in the sediments (greater than 4 em depth) where sulfate is minimized through methane oxidation. By using proxy indicators of the sediment depth of formation and availability of sulfate, this hypothesis was tested . These proxy indicators include porosity, availability of oxygen and sulfate, and methane concentrations as either free gas or hydrates . Porosity estimates were determined through petrologic examination of samples (optical and scanning electron microscope) and estimates of porosity replacement were determined by carbonate cement. Proxy evidence of the availability of oxidants was determined through petrologic examination of reduced or oxidized mineral phases (i . e. pyrite, glauconite, and hematite) . Finally, the influence of deep methane hydrates over seawater was established through stable isotopic measurements of o13C and o180.

PAGE 13

5 44 40' N 44 35' N 0 KM Figure 1 1 . Stud y area of Hydrate Ridge , located off the coast of Oregon. Carbonate Samples collected in this study came from the northern portion of Hydrate Ridge . Pockmarks (not shown) are located on the east side of the Ridge.

PAGE 14

6 0 3 5 10 15 • Aragonite .. I 3.5 .1: c Calcite . ,. • E 3.0 • 0 e 2.5 • .. • a: 0 . fS' 0 • •• 0 • ... • 0 1.5 o I""'" .....--oO Figure 1-2. Precipitation rates of aragonite versus calcite at 5 C. Aragonite is slightly favored at these temperatures, as shown in this figure. However, the addition of sulfate inhibits the precipitation of calcite . (Modified from Burton, E., L. Walter. 1987. Relative Precipitation Rates of Aragonite and Mg Calcite from Seawater: Temperature or Carbonate Ion Control? Geology, v .15. 111114 . Figure 1, pg. 112)

PAGE 15

CHAPTER2 HYDRATE RIDGE STUDY AREA Hydrate Ridge is an elongated anticline bisected by thrust faults associated with subduction . The study area is located 102 kilometers off the coast of Oregon ( 16 km from the Cascad i a deformation front) on the accretionary prism, at which ODP borehole site 892B is located (Figure 1-1 ). High organic content in the sediment column has caused methane venting in this area for at least the past 21 to 24 ky (Kastner et al. 2000) . Hydrate Ridge is named for the prolific hydrate formation in the prism sediments . Typical water temperatures on the sea floor of Hydrate Ridge are approximately 4.3C (Kastner et al . in press) . Estimated sedimentation rate of hemipelagic and terrigenous material is 2 cmlky, and sediments are general} y comprised of interbedded glauconitic quartz sands and silty clays (Ritger et al . 1987 , Schluter et al . 1998 , Westbrook et al . 1994). . Extensive carbonates , often associated with fault expressions on the surface of the seabed, are scattered around Hydrate Ridge. Massive (3-5 meters high) aragonitic chemoherms , carbonate build-ups precipitating from chemical reactions between methane rich fluids from depth and seawater, are outcropping at the seabed (Bohrmann et al . 1998). These chemoherms are frequently associated with fluid expulsion from faults , and can be identified through GLORIA side scan imagery (Carson et al . 1991 ). Chemoherms are often located at the top of the ridge, while muddy bottoms occur at the lower slopes of the ridge . Large boulder fields are located inside the pockmarks , and 7

PAGE 16

8 occasionally on the lower slopes away from the ridge . Along with authigenic carbonates, methane-related clam communities, bacterial mats, and bubbles of methane gas confirm active venting on Hydrate Ridge (Westbrook et al. 1994). ....

PAGE 17

CHAPTER3 LITERATURE SURVEY (BACKGROUND INFORMATION) The discovery of gas hydrates off the coast of Oregon was first noted by Kulm et al. (1986). Hydrates from Hydrate Ridge were then collected and described by Bohrmann et al. (1998). He noted the presence of carbonates near the surface on Hydrate Ridge, (700 meters) and bottom temperature ( 4 C) . Kastner et al. ( 1998) noted floating hydrates and methane plumes in the water column over Hydrate Ridge caused by the significant amounts of gas hydrates expelled from the sediment column . Sediment geochemistry of Hydrate Ridge was analyzed in ODP logs and results ( 1994 ). The logs documented high concentrations of H 2 S in the sediment column from near-surface to 60 meters depth . The results also concluded that bacterially-mediated methanogenisis occurs in the upper part of the sediment column (p r obably upper 2 meters). In the upper 20 meters, pore water cr, Ca2 + , Na+, and alkalinity varied slightly with the dissassociation of the methane hydrates. Below 20 meters , decreases in Cl, Na+, and Mg2 + concentrations and alkalinity are possibly the cause of advecting fluids (Westbrook et al. 1994 ) . Schluter et al . ( 1998) sampled fluids from the borehole, and found that 02, N03-, Si+4, H2S, sol+, and cr are all compositionally close to bottom water . Methane was considerably atypical, ranging between 0.2-3.5 mM, which is 106 times greater than seawater concentrations (Schulter et al . 1998) . The sediments drilled from the borehole were classified as Pliocene terrigenous silty clays and clayey silts with some sand layers and Pliocene turbidites (West brook et al . 1994 ) . 9

PAGE 18

10 Mineralogy Global observations of carbonates associated with methane consistently describe three mineralogies: aragonite, high Mg calcite, and dolomite (Jorgensen, 1990, Kastner et al. 2000, Martin et al . 1996, Paull et al. 1992, Rad et al.1996, Suess et al. 1999). Ritger et al. ( 1987) examined dredged and Alvin-collected samples from the lower Cascadia continental slope in thin section and SEM for mineralogies and crystal morphologies, which included radiating aragonite crystals and detritus-rich micritic calcite cement. Ritger et al . ( 1987) also found that all high Mg calcite and dolomite samples were comprised of microcrystalline cement. They found cubic or framboidal pyrite in every thin section they analyzed . In thin section, aragonite primarily formed as " elongate, radiating crystals" (Ritger et al . 1987). Kulm and Suess ( 1990) also observed samples from the Cascadia continental slope that contain patchy pure aragonite and a micritic cement mixture of silt and clay. They found carbonates that were comprised of well-sorted glauconitic grains, with some pyrite. This dark glauconitic matrix was cemented by a combination of aragonite and high magnesium calcite, or by pure aragonite (Kulm and Suess 1990). Bohrmann et al. (1998) noted the pure botryoidal and isopachous aragonite in samples collected from Hydrate Ridge, and hypothesized that the crystals grow in gas hydrate cavities. The aragonite needles were 2-10 Jlm thick, with botryoidal radii of 3-15 mm or layers of 40 to 600 Jlm thick (Bohrmann et al. 1998) . Sample and Reid ( 1998) described high Mg calcite and dolomite samples collected by Alvin from the Cascadia accretionary wedge with multiple veins, and occasionally floating detrital grains located in the veins. They also noted minor amounts (0-5%) of

PAGE 19

11 unbroken foraminifera tests, which could indicate a shallow cementation process (Sample and Reid 1998). Utilizing X-ray diffraction (XRD) to determine mineralogy, Ritger et al. (1987) found dominantly high magnesium calcite cement which contained 6-23 mol% MgC03, with an average of 12 mol% MgC03 . This average was much higher than normal deepwater calcites, which contained less than 4 mol% MgC03. Ritger et al. suggested this could point to a precipitation at higher temperatures (Ritger et al . 1987) . Greinert et al. (2001) used XRD analysis to classify carbonates from Hydrate Ridge into different phases based on amounts of magnesium: aragonite, low Mg calcite (less than 8 Mol% MgC03), high magnesium calcite (8-20 Mol% MgC03 ), protodolomite (30-40 Mol% MgC03 ), and dolomite (40-55 Mol% MgC03 ). In a study of gas hydrates in the sediment, Bohrmann et al. ( 1998) recognized two types of carbonates directly related to shallow methane hydrate deposits, including aragonite (most common) and high Mg calcite (14-19 Mol% MgC03 ). Kopf et al. ( 1995) found that 80% of their borehole, Resolution-collected samples from the Cascadia accretionary wedge were composed of high magnesium calcite, while the remaining 20% were aragonite and dolomite . Sample and Reid ( 1998) used XRD to identify minor constituents in fine-grained carbonate samples. Secondary minerals found in XRD analysis of authigenic carbonate samples from the Cascadia margin included quartz, plagioclase, clay minerals, pyrite, and glauconite (Sample and Reid 1998) . Ritger et al. (1987) analyzed their Cascadia samples for the total mass percent carbonate through an acid-leach weight-loss procedure, and the samples grouped into two categories. The first were low-carbonate samples, with less than 20% CaC03 , while the second had much higher amounts of carbonate, between 25% and 90% CaC03 . Sample

PAGE 20

12 and Kopf ( 1995) analyzed their high Mg calcite samples from off the coast of Oregon for total percent carbonate, and found a range between 1% and 10% CaC03 . A second area analyzed had higher carbonate content (up to 25% CaC03 ). Total percent carbonate did not have any clear correlation with pore fluid chemistry, or Mg/Ca ratios (Sample and Kopf 1995). Morphologies On a global basis, the morphologies of authigenic carbonates are also similar. Carbonate chemoherms, crusts, slabs, boulders, and chimneys are common across the globe, and Hydrate Ridge is no exception (Kastner et al. 2000, Jorgensen 1990, Martinet al. 1996, Paull et al. 1992, Rad et al. 1996, Suess et al. 1999). Figures 3-6 show examples of chemoherms, crusts, slabs, and boulders derived through Alvin dives on Hydrate Ridge and in hand samples. Ritger et al. (1987) found Hydrate Ridge breccias that included mudstones, sandstones, and conglomerates and were often cemented by magnesian calcite. Greinert et al. (2000) petrographically categorized various authigenic carbonate phases from Hydrate Ridge as homogeneous mudstones, tectonized mudstones, bioturbated mudstones, mudclast breccias, intraformational breccias (clasts of intraclasts, extraclasts, shells, and mudclasts ), cemented bioturbation trails, and gas hydrate-associated carbonates. Mudstones in general were cemented by dolomite to protodolomite and were rich in pyrite. The mudclast breccias were grain to matrix supported, with areas of centimeter-thick bands of pure radial aragonite crystals. The irregularly shaped intraformational breccias included intraclasts (sometimes glauconitic), extraclasts, bioturbation casts, shells, and mudclasts . Both the clasts and the matrix contained pyrite, and the cement was primarily aragonite needles. Greinert et al. (2000) divided

PAGE 21

13 gas-hydrate related carbonates into aragonitic collapse breccias and elongate pores (gas hydrate relic). The collapse breccias were grain supported and cemented by boytryoidal aragonite . The elongate-pore carbonates exhibited distinctive megapores, which resemble gas hydrate layers (Greinert et al. 2000). Factors Influencing Carbonate Precipitation Extensive research has been published on the factors influencing the precipitation of authigenic methane-related cold-water carbonates , particularly from the accretionary prism off Oregon . Ritger et al. ( 1987) presented hypotheses on cementation processes and the conditions that lead to carbonate precipitation based on samples collected from the lower continental slope of Cascadia . They believed that cementation was induced by increased alkalinity associated with sulfate reduction, abundant availability of Mg2+ and Ca2+ ions from seawater, and decreased C02 solubility due to a pressure decrease. In anoxic conditions, methane is oxidized and sulfate reduced by the reaction: + S042 HS+ HC03+ H20 (Ritger et al. 1987) At the oxic seabed surface, the methane is oxidized by the reaction: (Ritger et al. 1987) The calcium that combines with the C03 2 to enable precipitation of carbonates is from seawater . Ritger et al . ( 1987) believed that the aragonite was younger than the calcite because aragonite needles crystallized on the calcite cement. One important question that many studies have addressed is how so many different types of carbonate (aragonite , high magnesium calcite, and dolomite) are found in the same area. To answer this question, an understanding the conditions under which aragonite and calcite precipitate is necessary . Burton and Walter ( 1987) demonstrated that

PAGE 22

14 the primary controls on carbonate phase precipitated are temperature and carbonate saturation state of seawater (Figure 1-2). At 5C (the average water temperature at Hydrate Ridge), rates of calcite precipitation are equivalent to those of aragonite regardless of saturation state, but as temperature increases, aragonite is the favored precipitate. However, an important control on the preferred phase is dissolved sulfate (Burton and Walter 1987). In the zone of sulfate reduction, high magnesium calcite is the favored phase, while in the sulfate-rich pore waters near the surface, aragonite is the favored form. In laboratory experiments, the presence of dissolved sulfate strongly inhibits the precipitation of calcite and dolomite. The reduction of S04 2 -assists dolomitization by removing the S04 2 inhibitor, increasing the alkalinity, and producing NHt that can exchange with Mg2+ and free the magnesium for dolomitization (Baker and Kastner 1981 ). Morse et al. ( 1997) studied the importance of temperature and Mg:Ca ratio in pore water on precipitation of calcite and aragonite. They found that both factors influenced the type of precipitation, but that minor temperature changes dramatically effected the Mg:Ca ratios. Higher temperatures (+6C) and normal seawater Mg:Ca ratios (5 :1) favored aragonite precipitation, while low temperatures ( <6C) and a Mg:Ca ratio less than 1 :5 favored calcite formation. Therefore, Morse et al. also suggested that paleoclimate/temperature could greatly effect carbonate phases precipitated, and should be taken into account. However, wide changes in temperatures are not likely to occur on Hydrate Ridge due to the depth and near-zero temperatures. Dolomite formation in methane-seep environments is more difficult to explain than aragonite and high Mg calcite . Some authors argue for direct dolomite precipitation, while other argue for a recrystallization of calcite to dolomite over time. Baker and

PAGE 23

15 Kastner ( 1981) suggested the controls on dolomitization include not just the Mg/Ca ratio (as previously believed), but also the concentration of sulfate in the formational waters . Teal et al. (2000) also emphasized the importance of microbial sulfate reduction, along with methanogenisis. They suggested that dolomite precipitation was bacterially mediated at normal seawater salinity in anoxic conditions promoted by methanogenisis . They also found that the dolomitization was not obviously correlated with depth , location, sediment texture, porosity, or mol%MgC03 • Dolomitization in this model would be accompanied by authigenic pyrite precipitation. They also supported a recrystallization theory (calcite to dolomite) through burial diagenisis under some conditions on the seafloor (Teal et al. 2000). A paper by Baker and Bums (1985) covered the occurrence of dolomite on continental margins . In areas similar to Hydrate Ridge, they concluded that primary controls on dolomitization were the availability of calcium and the concentration of sulfate. Similar to other published papers, they promoted a theory of dolomitization in the upper 20-30 meters of the seabed in the zone of sulfate reduction. This depth of precipitation is also supported by the availability of magnesium, which decreases more slowly with depth than calcium. Therefore, dolomite can precipitate at sediment depths greater than calcite. The sources for magnesium are varied , and may include calcite, clays, and diffusion from seawater. Finally, Baker and Bums also proposed the primary input of magnesium was from diffusion, and therefore the dolomite must form in the upper few tens of meters of the seabed (in the zone of sulfate reduction) (Baker and Bums, 1985).

PAGE 24

16 Stable Isotopes Oxygen and carbon isotopic analyses have been powerful tools in the study of authigenic carbonates because o13C and 0180 values can reflect methane-related carbon source, and can constrain porewater compositions and temperature at formation, respectively. o13C values can reflect the methane source (biogenic or thermogenic), and the influence of seawater during precipitation. Typical 013C values for seawater are around O%o PDB, while biogenic sources are highly depleted, recording o13C values of -15%o to -90%o PDB, and thermogenic sources record o13C values of -30%o to -50%o PDB. However, mixing of seawater and biogenic methane can increase the o13C values to the thermogenic range. K venvolden ( 1995) suggested that when methane was the dominant gas hydrate, then the 013C values were most likely due to microbial methane . The 013C values of carbonates from Hydrate Ridge were analyzed for the influence of hydrates, methane gas, and seawater during precipitation (Kastner et al. 2000, et al. 1999). 0180 values of carbonates are controlled primarily by mineralogy, temperature, and the oxygen isotopic values of the water from which the carbonate precipitated. It is important to note here that oxygen isotopic values can be effected by climatic changes during the Cenozoic (through temperature changes), but these changes are not significant on Hydrate Ridge in the past 20 kyrs (Hudson and Anderson 1989). Most 0180 values from Cascadia are primarily affected by the isotopically-heavy waters (with respect to seawater) which are produced by methane hydrates (+O.l%o to +7.9%o). Methane-rich waters are generally isotopically heavy, and will affect the 0180 values for precipitating carbonate (Matsumoto 1989).

PAGE 25

17 Ritger et al. separated their isotopic analyses into low carbonate and high carbonate samples . The low carbonate samples (less than 20% CaC03) had carbon isotopic values of -0% PDB, but had unusually low o180 values (-1.6%o to -13%o PDB). The high carbonate (over 25% CaC03 ) samples reflected a methane influence as seen in o13C values that ranged from -66.1%o to -34 .9%o PDB . Positive o180 values (+2.78%o to +8.27%o PDB) in the high carbonate samples reflected "a more complex signal than just temperature of formation" (Ritger et al. 1987). Ritger et al . believed in a biogenic methane source because the measured carbon isotopic values were fairly depleted in 13C. Sample et al. measured a range of -1%o to -25%o PDB for carbon isotopes from samples collected by Alvin along the frontal thrust in Cascadia, which were assumed to reflect a thermogenic source. The measured oxygen isotope values were extremely depleted (-4%o to -25%o PDB), indicating high temperatures (H>0C) during precipitation (Sample et al . 1993). The isotope data for the carbonate cements trended along a line from depleted in o180 and average in o13C to enriched in o180 and depleted in o13C. Sample et al . hypothesized that the isotope values were a direct result of warm, deep fluids ( 1 00C) migrating up through faults and precipitating carbonates at shallow depths (Sample et al. 1993). Greinert et al. (2000) reported the widest range for carbon isotopic data occurred in chemoherm samples (+26.2%o PDB to -56%o PDB). These samples were then divided into six groups based on lithology, minerology, and carbon isotopic data (see previous discussion for lithologies) . The dolomite mudstones comprised the most positive o13C values, 10%o to 26%o PDB . Greinert et al . (2000) proposed these positive values were a result of the reduction of C02 in methanogenic zones . A second group, consisting of

PAGE 26

18 mudclast breccias, mudstones, and cemented bioturbation trails, had carbon and oxygen values ranging from, -34.6%o to -46 .7%o PDB and 6 .3%o to 1.5%o PDB respectively. The high magnesium calcites from the next group had o13C values up to -29.2%o PDB and 0180 values around 4.3%o PDB. The majority of the samples fell into the last category, composed primarily of intraformational breccias and mudclasts. Their carbon isotopic values ranged between -38.2%o and -55.2%o PDB . Aragonite from this group had 0180 values between O%o and -0.3%o PDB (in equilibrium with ambient seawater), while the high magnesium calcite 0180 values fell between 4 .9%o and 7.3%o PDB (in equilibrium with isotopically heavier water) . They concluded that high magnesium calcite precipitated under different conditions than the aragonite, and were more influenced by gas hydrates reflected in the positive o180 values (Greinert et al. 2000). Bohnnann et al. ( 1998) found two isotopically distinct end members in a study of carbonates collected from Hydrate Ridge. These results, which are similar to Greinert et al. (2000), separated a hydrate-influenced high magnesium calcite and an ambient-pore water-influenced aragonite. The recorded carbon isotope values for the high magnesium calcite and aragonite were between -40.6%o and -54.2%o PDB , reflecting a methane influence in both phases. The aragonite oxygen isotopes averaged 3.7%o +1-O.l%o PDB, while the magnesium calcite averaged 4.9%o +1-0.1%o PDB. The aragonite oxygen isotope value differed from SMOW by only 0 . 1 %o, while the high magnesium calcite differed by 1 %o (Bohnnann et al . 1998).

PAGE 27

CHAPTER4 METHODS Collection of samples and push cores took place over a period of three summers in 1998, 1999, and 2000 . DSRV Alvin collected carbonate hand samples and push cores, which were then returned to Lehigh University and the University of Florida for analyses. At Lehigh, cobbleand boulder-sized samples were cut with a rock saw into hand-sized specimens for thin section preparation and to study macroscopic texture and mineralogy (Figures 7-11 ) . A total of 59 thin sections were produced by Texas Petro graphics. These thin sections were analyzed and photographed through the use of a Zeiss photomicroscope ill, primarily at x2.5 and x10 settings. X-ray diffraction analysis was run on 20 samples at the University of Florida, all from push cores taken by Alvin . The cores ranged in length (10 em and 18 em), and were subsampled in two centimeter intervals. The intervals were then sieved into 4 size fractions, <-10, >-2.750. The largest fraction, >2. 7 50, was photographed and then ground to 200Jl . Some of the clasts had obvious zoning, and three were drilled for zones (6 samples). All two-centimeter-interval samples (14 samples) and zoned clasts (6 samples) from the cores were run on a Phillips APD 3600 XRG 3100 generator at 45kV/30mA. They were run with a step size of .02, count time of 2 seconds, and 29 (peak angle range) of 24-38. The samples were mixed with an internal standard, corundum, at approximately 50/50 mass. Data were collected and interpreted using Jade 3.1+ software. XRD was also run at Lehigh University on 44 19

PAGE 28

20 samples drilled from carbonate hand specimens . The generator and settings were the same as above . High Mg calcite samples were rerun ( 18 samples) to identify multiple peaks in the calcite-dolomite range (29.5-31 o 2S), and qualitatively distinguish high Mg calcite from calcite and dolomite . The position of the peak in the calcite-dolomite range determined the mol% MgC03 , based on Goldsmith et al . (1961). Poorly ordered dolomite peak was assumed to lie at 30 . 25 2e. The software used at Lehigh was Phillips APD1700 . Total mass percent carbonate was calculated using a model 5011 C02 coulometer at the University of Florida . The software used to analyze and interpret the percent carbon was WinCarb. Thirty-seven samples were drilled from carbonate hand samples using a Dremel hand drill, and were weighed and analyzed. The samples were chosen from a variety of different lithologies, colors, clasts , and mineralogies. Some transects along a hand sample were drilled in order to study changes from rims to interiors. Clasts sieved from the push cores were also selected to be analyzed for total mass percent carbonate . Five clasts (>-1.750) from two intervals, 2-4 em and 8-10 em, were chosen as a representative sample and photographed. These were then ground , weighed, and analyzed . The percent carbon calculated by the coulometer was then multiplied by the known standard value of 8.33 to find total percent carbonate. The total weight percent carbonate was then converted to volume percent carbonate through the equation: Vol% Carbonate= (wt.% carb./2 . 90g cm-3)/((wt. % carb./2 . 90g cm-3)+(wt.% noncarb./2.65 g cm-3)) x 100 (after Scamman 1981) Carbon and oxygen isotopic values were measured at both the University of Florida and Lehigh University . The samples were ground by hand to a fine powder, and

PAGE 29

21 then weighed. The isotopic values were measured by a Kiel III Carbonate Prep Device coupled to a Finnagan-MAT 252 isotope ratio mass spectrometer located at the University of Florida . A total of 101 samples were run at Florida, and 19 at Lehigh. XRD samples from the push cores were not run for isotopic values because the mixed mineralogy found by the XRD analysis would have greatly complicated the interpretation of the isotopic results. Each sample was evaluated for the 0180 value of formation waters given a temperature of 3 . 9C using standard equations for their respective mineralogies (aragonite and calcite) through the following equations : for aragonite (PDB) (3.52(sqrt(3 . 522-(4*0.03*(19-T)))/(2*0 . 03) (Grossman and Ku 1981), and for Mg-calcite (PDB) 2.78(1Q6y2)-2.89+0.06 mol%MgC03 (Tarutani et al . 1969) Scanning electron microscope analysis (SEM) was performed on six samples that represented the mineralogies present on Hydrate Ridge. The samples were trimmed to the appropriate size, sanded on their base to provide a flat surface (tops were left with natural texture), and then mounted using silver glue to stubs . The samples were then coated with iron and analyzed by a Joel JSM 6400 scanning microscope. Photographs were digitally recorded as 200 dp TIFF images . Individual grain mineralogies of SEM samples were determined through EDS (energy dispersive spectrometry), using the presence ofMg, Ca, C, and 0 peaks as an indicator of calcite or dolomite, and Sr, Ca, C, and 0 peaks as an indicator of aragonite.

PAGE 30

CHAPTERS RESULTS Previous field studies of Hydrate Ridge over the past 15 years revealed that authigenic carbonates occurred in several different morphologies which have been discussed previously (Figures 5-1 through 5-9) . Chemoherms are large, porous mounds of carbonate (typically acicular aragonite) which are located at and above the seabed (Sample et al . 1993) . The crusts are thin (1-2 em) outcrops of carbonates jutting from the seabed, whereas the slabs are similar but thicker, (2-4 em) less porous, and often lie in stacks . Crusts are generally composed of aragonite, while slabs are typically high-Mg calcite (see below) . Mineralogies The mineralogies of samples were measured with XRD analysis (Table A-1). The XRD was performed on sections of 20 hand samples to identify mineralogy . The hand samples measured included aragonitic chemoherm samples, dolomitic samples, high magnesium calcite samples, and breccia samples. In these hand samples, multiple carbonate mineralogies did not occur in any one sample except for in the breccias (see figure 5-10 for example ofXRD analyses) . The different carbonates of chemoherm samples (Figure 5-5) were analyzed by XRD, and found to be aragonite and quartz (Table A-1) . Ten samples from crusts (Figure 5-6) were tested with XRD and found also to contain aragonite and quartz (Table A-1) . Carbonate slabs (Figure 5-7) collected from the seabed surface of Hydrate Ridge had been identified in the literature as containing high 22

PAGE 31

23 Mg calcite. This observation was confirmed through XRD analysis of slabs from Hydrate Ridge (Appendix Table 1 ). Slab samples were composed of high Mg calcite, containing approximately 6 to 11 mol%Mg. The boulder samples were proto-dolomite and dolomite , and difficult to distinguish in XRD from very high Mg calcite (Figure 5-8, Appendix Table 1). Almost all dolomitic samples were disordered, with 39-42 mol%Mg . As stated in the methods section, this study considered any peak reflecting a spacing . over 3.25 ' angstroms (30 . 25 28) to be highly disordered dolomite, and therefore these samples fell under protodolomite category (not high Mg calcite) . The breccias were more complicated than the other carbonates collected at Hydrate Ridge (Figure 5-9) . They contained multiple mineralogies and lithologies in their clasts and cements . From XRD analysis, clasts consisted of feldspar and quartz , pure calcite, high Mg calcite, dolomite, and, less commonly, aragonite . However, aragonite was the dominant cement (Figure 5-9) seen in the breccias as determined by XRD. Some of the breccias had very similar clasts with the same mineralogy, but others had a wide variety of clast sizes and mineralogies . The high Mg calcite clasts ranged between 9 and 18 mol%MgC03. Finally , XRD was run on the largest, authigenic , hand-picked clasts from the bulk sediments sieved in the two push cores (sieved size -2 .750). One sample from each two-centimeter interval was analyzed by XRD, and results can be seen in Figures 5-11 and 5-12 . The first core analyzed (B20) was 10 em long, and every interval has mixed mineralogies. Aragonite was present in all intervals except for 6-8 em. High Mg calcite was also present in all intervals, but the peaks were wide for all intervals, implying a range of mol%MgC03 in the calcite . Pure calcite was present only in the 4-6 em interval. . . .. .

PAGE 32

24 Dolomite was also present in all intervals, and was all proto-dolomite, except for a double peak in the 4-6 em interval of 50% near-stoichiometric dolomite (31 o 28) and 50% proto-dolomite. The dolomite became more disordered with depth in the core, lying closer to 30 28 . The next core (CJ4) was longer, at 18 em (Figure 5-12). Aragonite was present in the top 14 centimeters, below which it disappeared . High Mg calcite was not present in the top two centimeters or the bottom 6 em , but was present from 2 to 12 em depth . The high-Mg calcite was concentrated in the center of the core. A small peak of low Mg calcite was noted in the upper two centimeters of CJ4 . Dolomite was not present in the top two centimeters . The dolomite was primarily disordered, especially in the 6-8 em interval and was more disordered in the center of the core. Both cores analyzed contained a detrital quartz and feldspar component as seen in XRD. Due to the mixed mineralogy results from XRD analysis of the cores, some of the other large clasts from the same intervals in the cores as the clasts analyzed in XRD were drilled from brecciated sections or concentric zonations (examples can be seen in Figures 5-13 and 5-14 ). Two clasts with zones were chosen from CJ4 interval 12-14 em. One of the clasts analyzed resulted in no carbonate found by XRD in either zones . Only a quartz peak was present in the diffractogram for both zones, so the clast apparently was comprised entirely of detrital material. The second clast from CJ4 12-14 em had an interior zone consisting of dolomite with minor amounts of quartz . The exterior was also primarily dolomite, but had a peak of low Mg calcite. The final clast analyzed was taken from CJ4 interval 14-16 em. The interior and exterior zones were composed of dolomite and minor quartz, and no other types of carbonate appeared to be present.

PAGE 33

25 Porosity The porosity of the carbonates collected from Hydrate Ridge was analyzed in thin section, SEM, and as total percent carbonate assuming complete filling of the pore spaces by authigenic carbonate. The mass percent calcium carbonate was calculated for many of the same samples also run for oxygen and carbon isotopes (Figure 5-15). Chemoherm samples had an average of 73% calcium carbonate, while the aragonitic crust samples had the highest average of 82% CaC03 • High Mg calcite samples had an average of 74% CaC03 , and dolomitic samples averaged at 61% CaC03 • Aragonitic samples had a much smaller range of percent carbonate values with a standard deviation of 9% than the high Mg calcite (standard deviation of 14%) or the dolomite (standard deviation of 28% ). All percent carbonate values are listed in Table A-2. Total percent carbonate by weight was also converted to volume percent carbonate, which estimates the minimum initial porosity in the sediments before precipitation and replacement of pore volume with carbonate cement. This calculation implied the same trends in porosity noted above, with the highest porosities in aragonite (55%-90%) and the lowest in dolomite samples (42%-84%) (Table 5-1). Volume percent calcium carbonate was also run on five small clasts from two intervals in two cores, B20 and CJ4 (Figure 5-16). Interval 0-2 em in B20 had an average of 75.6% CaC03 , which then dropped to near-zero averages in the 2-4 em and 4-6 em intervals. The 6-8 em and 8 10 em intervals increased back to 70-80%. Core CJ4 intervals primarily ranged between 65 and 83, with three exceptions. Intervals 6-8 em and 10-12 em had lower volume percent CaC03 values at 46% and 51% respectively, while interval14-16 em dropped to 17%.

PAGE 34

26 Finally , larger clasts from every interval in each core were analyzed for total percent calcium carbonate. The percent carbonate for core B20 was fairly consistent, with a range between 71% and 78% CaC03 , except for interval 2-4 em which had almost no carbonate (0 .1% CaC03 ). The second core, CJ4, had an average of 63% CaC03, with a range from 18% to 84% CaC03 . No significant trends with depth were detected in the percent carbonate . Porosity can also be determined qualitatively through thin section and SEM analyses by visually approximating pore space . Petrologic examination of carbonate samples from Cascadia revealed great variability in fabric and can be seen in Figures 19-42 . The chemoherm samples had two distinctive colors in the matrix (dark grey and light tanFigure 5-5), as well as white aragonite needles that grew into pore spaces . The chemoherm samples had the highest percentage of void spaces of all samples studied. These void spaces contained elongated acicular aragonite that often interlaced (Figures 5-17, 5-18, and 5-19) . Banding in the aragonite seemed to be a dissolution feature, where the needles had dissolved and then precipitation resumed (Figures 5-20 and 5-21) . The needles tended to grow in fan shapes "rooted" in the darker micritic matrix; the boundary between the needles and the dark matrix was pronounced (Figure 5-17). Benthic foraminifera were plentiful in the chemoherm samples, and were usually whole. Aragonitic samples from crusts collected from the seafloor had a similar structure to chemoherm aragonite samples, but contained fewer, shorter aragonite needles and more micritic aragonite than the chemoherm samples in SEM analysis (Figure 5-22). Clamshells were often incorporated into the matrix (Figure 5-23), and dissolution features on the shells were present. These samples also had high amounts of porosity . Whole

PAGE 35

27 forams and glauconite were common . The matrix was composed of distinct dark grey and light tan sections (Figure 5-24) which were the same basic mineralogy based on XRD analysis . Clay and detrital material was common in both colored matrixes, and could be seen in SEM (Figure 5-25) . The boundary between these different matrixes varied between both distinct and diffuse , depending on the sample . Porosity decreased in the grey matrix when compared to the tan . Some structures such as veins (some infilled with aragonite) and voids that were parallel to shells were present in the crusts. One thin section had graded beds of clasts (similar to Figure 5-30). Samples from a high magnesium carbonate slab were analyzed in thin section and SEM (Figures 5-26 through 5-29) . These samples were much less porous, with no acicular aragonite present (Figures 5-26 and 5-27) . The slab matrix in hand sample was a homogenous, medium to well indurated, micritic calcite (light grey in color), and had a thin darker-colored rind . The high magnesium calcite was all micritic, and therefore was not easily analyzed in thin section (Figure 5-26) . These samples had much more detrital material present than in the aragonite samples (Figures 5-28, 5-29, and 5-30). Forams were present in all the thin sections, usually with a 50/50 mix of fragmented and whole . One hand sample had only fragmented forams . Dolomite samples were collected from the surface in push cores and as boulders , often occurring in boulder fields. Their matrix was homogenous , typically light grey colored (similar to the high Mg samples), and highly indurated . As stated in the methods section, this study considered any 28 peak over 30.25 28 to be highly disordered dolomite, and therefore these samples fell under the proto-dolomite category . The dolomite samples had very low porosity (Figure 5-31 through 5-33). Fragmented forams

PAGE 36

28 were present in every thin section. Subparallel veins similar to the ones in the high Mg calcite (Figure 5-30) were present in these dolomite samples , and they were occasionally infilled with acicular aragonite . The veins with aragonite tended to be located near the edges of the samples, possibly indicating secondary precipitation. In SEM images, both veins and micritic dolomite looked identical, with similar mineralogy and grain size in both (Figure 5-33). These dolomite samples had larger forams compared to previously described thin sections of calcite and aragonite . Breccias are common on Hydrate Ridge, and twelve representative breccias were chosen for thin sectioning (Figures 5-34 through 5-36). Only two of the breccias analyzed contained more than a few forams, and those were primarily comprised of whole, large forams (Figure 5-36). These samples had various percentages of porosity, ranging generally from 10% to 60%, determined by estimation in thin section . Large fragments of bivalves (15 em in length) were common in breccias , and the cement between clasts was typically acicular aragonite . As stated previously , the breccias were a mixture of mineralogically-similar clasts and distinct , mixed-mineralogy clasts, typically matrix supported. Some clast boundaries were diffuse, while others were sharp. The clasts were subrounded to subangular, a variety of sizes , and were a mix of detrital and carbonate. Structure was present in the breccias in the form of veins and elongated pores seen in the clasts. Oxidant Availability as Proxy for Depth in Sediments Oxidant availability in the carbonate samples from Hydrate Ridge also show wide variability . The presence of pyrite and glauconite indicate reducing conditions, while hematite staining implies oxidizing conditions. The chemoherm samples contained pyrite,

PAGE 37

29 generally rounded (framboidal) in shape , and no visually-detected hematite staining present in any of the thin sections . Noticeably little glauconite (seen in thin section as clusters of nodules) was present in the thin sections. In crust samples, framboidal pyrite was common, and hematite staining was absent. The samples from high Mg calcite slabs contained pyrite that was present in larger amounts than in the aragonitic samples, primarily occurring as infilling veins or voids in the matrix (Figure 5-37) , but also as framboidal clusters . A small amount of the pyrite and surrounding areas were stained with hematite in one sample, while the other had no evidence of staining . Very little glauconite was present in any of the thin sections. Structures such as parallel veins and cracks were common in these samples (Figure 5-38), with glauconite infilling veins in one thin section . Dissolution rinds were seen in many samples in thin sections (Figure 5-39) . Dolomite had framboidal and infilling pyrite present in thin section and SEM (Figure 5-40). The framboidal pyrite was larger than any seen in previous samples (>5 mm). Hematite staining was common in these samples, and was concentrated near veins or cracks . The presence of glauconite was rare in these thin sections. The samples also commonly contained a dark red weathering rind (2-4 mm thick), possibly composed of iron and manganese. Every breccia had small amounts of pyrite present. The pyrite was present as both framboidal clusters and within veins, often in the same thin section. The amount of hematite staining varied with individual breccias, ranging from no stains to stains commonly occurring particularly concentrated in veins . The amount of glauconite was

PAGE 38

30 highly varied in these thin sections. High amounts of glauconite and the presence of forams tended to correlate well in the breccias . Stable Isotopic Evaluation Stable isotope analysis was run on all hand samples used in thin section and XRD but not core samples (see Appendix Table 1). o13C values for chemoherm samples ranged between -40.9%o and -48 .4%o PDB. The 3180 values for the chemoherm samples plotted between 3.8%o and 4 .6%o PDB. The aragonitic crust samples were analyzed, and the carbon isotope ranged between -41.9%o and -49.9%o PDB (Figures 5-41 and 5-42). Four unusually high points also occurred in the carbon isotope data for these samples, falling between -9.8%o and -0.5%o PDB (Figure 5-42). The oxygen isotope values for aragonitic crusts plotted between 4.1%o and 5.1%o PDB, with no significant outliers. High Mg calcite o13C values ranged between -52.0%o and -30.0%o PDB, while the oxygen isotopic values were all positive (Figure 5-43). For two high Mg calcite slabs, samples from microdrill transects were analyzed (Figure 5-44), and the o13C values ranged between -30.5%o and -44.8%o PDB. The oxygen isotope values fell within 3.7%o and 5.4%o PDB. Figure 5-44 transects depicts more depleted o13C values towards the bottom of the slab. The oxygen values have no clear trends along the transect. Dolomitic samples were analyzed, and the carbon isotope values ranged between -34.1%o and -39.6%o PDB (Figure 5-45). One o13C outlier of +4% was noted in one of the dolomitic samples. The 3180 values for the dolomite fell within 5.5%o and 7.2%o PDB, except for one outlier of -41.0%o PDB (Figure 5-45). These two carbon and oxygen isotopic outliers from the dolomites were from the same sample, drilled from the interior of a dolomite boulder. Four samples from a dolomite breccia produced a range of o13C

PAGE 39

31 values that clustered in two areas, 21.6%o to 16.8%o PDB, and -21.2%o to -20.1 %o PDB. Similarly, the oxygen isotopic data clustered at 3.8%o to 4.0%o PDB, and 6 .7%o to 7.0%o PDB (respectively with the carbon clusters: not pictured) . The breccias contained a wide variety of isotopic compositions due to the mixed composition of the clasts (Figure 43 ). For cements, the broad range of carbon isotopic values fell between -34.4%o and -54 .7%o PDB, with outliers at -22 .5%o, -12.4%o, -1.2%o, and 15.0%o PDB . The oxygen isotopic values ranged between 2.9%o and 6.6%o PDB, with no notable outliers. The breccia clasts did not cluster together, with a broad range of carbon and oxygen isotopic values occurring in many samples ( -34.4%o to -45.0%o PDB and 3.0%o to 6.2%o PDB, respectively). One breccia contained all low Mg calcite clasts (determined through XRD), and the o13C and o180 for the clasts clustered around -35%o and 3%o PDB, respectively.

PAGE 40

32 Figure 5-l. Alvin photograph of the top of an aragonitic chemoherm from Hydrate Ridge . Field of view is approximately 8 meters. • . s

PAGE 41

33 Figure 5 -2. Alvin photograph of aragonitic crusts from Hydrate Ridge. Field of view is approximately 5 meters .

PAGE 42

34 Figure 5-3. Alvin photograph of high Mg calcite slabs from Hydrate Ridge. Field of view is approximately 4 meters.

PAGE 43

35 Ill Figure 5-4. Alvin photograph ofFe-Mn encrusted dolomite boulders (boulder field) from Hydrate Ridge. Field of view is approximately 10 meters.

PAGE 44

36 Figure 5-5. Hand sample (MSOl) of chemoherm with multiple colors of dark grey and light tan carbonate and acicular aragonite. Upper box shows location of thin-section sample in Figure 19. Lower left box shows " banding" in acicular aragonite (see Figure 22) .

PAGE 45

37 Figure 5-6. Hand samp l e (MS03) of aragonitic crust with m ul tip l e co l ors of dark gre y and tan cement. Lower box is location of thin section samp l e i n Figure 26 .

PAGE 46

38 Figure 5-7 . Hand sample (MS05) of high Mg calcite slab with thin weathering rind . Upper bo x indicates the location of thin section shown in Figure 28.

PAGE 47

39 Figure 5-8. Hand sample (MS07) of dolomite boulder with Fe -Mn weathering rind.

PAGE 48

40 Figure 5 9 . Hand sample (MS15) of breccia from Hydrate Ridge . Notice the sorting in parallel bands and the angularity of t h e clasts.

PAGE 49

41 Sample: j j 3 File: JJ3R31.SM I' 4-NOV-99 16: 21 x10 3 5.00 4.05 3 . 20 2.45 1.80 1.25 0.80 0 . 45 0.20 0 .05 24.0 ./ Corundum : ' Mg Corundum .. Corundum ,.,. Quartz Mg Calcite II II II 26. 0 28. 0 30. 0 32. 0 34.0 36.0 38. 0 Figure 5-10. Example ofXRD from drilled samples ofhand samples. Notice the presence of only one carbonate mineralogy (high Mg calcite). The sample was mixed with a reference mineral (corundum). Secondary detrital minerals include quartz and feldspar . The position of the high Mg calcite peak between 29 and 31 29 was used to determine the mol % of MgC03 •

PAGE 50

Aragonite / Mg C alcite I • 42 Dolomite / Aragonite I Corundum • • Figure 5-11. XRD analysis performed on core B20 . Two centimeter intervals begin at the top of the graph and continue downward . through the sediment column (0-2 em is represented by the top line while 8-10 em is represented by the bottom line). Large unlabeled . peaks are standard used when running these The stoichiometric dolomite peak falls at 31 degrees; most of these samples contain ' with minor amounts of stoichiometric dolomite . Notice all of these samples are comprised dominantly of dolomite and aragonite, with a Mg calcite constituent also present in every sample. The mixed mineralogy could be related to breccia clasts .

PAGE 51

Aragonite / M , g 43 . ., ... , .:. •• • • i •' , I '''. Dolomite . •! Figure 5-12. XRD results from core, CJ4 . The top line represents the uppermost interval (0-2 em), and each line below is a consecutive two centimeter interval. This core has dolomite at all intervals, with the addition of at least one extra mineralogy (aragonite , calcite , or Mg calcite) : Mg calcite and aragonite tended to decrease with depth , probably indicating the replacement of those minerals with dolomite .

PAGE 52

44 Figure 5 -13. Example of zoned clasts from core CJ4. These zones could be the reason for a v ariety of mineralogies present in one clast (as recorded by XRD).

PAGE 53

45 Figure 5-14. Example of zoned clasts from core B20. The clasts recovered from the cores were commonly breccias, as pictured in this sample.

PAGE 54

100 90 80 70 Cll Ill 60 c 0 .c ... Ill 50 u c Cll 40 Cll 0... 30 20 10 0 46 High Mg Calcite Sample# Figure 5-15 . Percent carbonate values (see Figures 7 and 8 for examples) plotted by mineralogy. The aragonite is broken into dark grey, tan , and white (i.e. , acicular) colors , and they generally have the highest total percent carbonate. The high Mg calcite samples (see Figure 9) span the broadest range of values , but the limited dolomite samples measured also had wide variety. Every sample measured had at least 39% carbonate.

PAGE 55

47 Table 5-l. Comparison of total mass percent carbonate with calculated volume percent carbonate . The calculated volume percent estimates the initial porosity present in the sediments . Notice that the aragonite samples have the highe s t initial porosity, and the dolomite samples have the widest variety in initial porosity . Carbonate Type Total Ma s s Percent Carbonate Volume Percent Ca r bonate Dark Grey Aragonite 57 55 Dark Grey Aragonite 79 77 Dark Grey Aragonite 86 85 Dark Grey Aragonite 79 77 Dark Grey Aragonite 76 75 Dark Grey Aragonite 77 75 Tan Aragonite 69 67 Tan Aragonite 78 76 Tan Aragonite 67 65 White Aragonite 91 90 White Aragonite 70 68 High Mg Calcite 72 70 High Mg Calcite 63 60 High Mg Calcite 69 67 High Mg Calcite 94 94 High Mg Calcite 88 8 7 High Mg Calcite 73 71 High Mg Calcite 54 52 High Mg Calcite 39 37 High Mg Calcite 83 82 High Mg Calcite 69 •.;,, 67 High Mg Calcite 70 68 High Mg Calcite '84 82 High Mg Calcite . , , . ;. '79 _!. 78 H i gh Mg Calcite 84 83 High Mg Calcite '44 t ' 42 •.. • • !fl , High Mg Calcite 54 51 H i gh Mg Calcite . 58 y 56 High Mg Calcite 70 68 High Mg Calcite 73 71 High Mg Calcite 76 74 Dolomite 85 84 Dolomite 45 42 Cores B2-4-1 62 Single Clasts 60 82-4-3 70 From Two 68 82-4 4 94 Intervals 94 82-4-5 76 75 BS-10-1 73 72 88-10-2 81 79 BS-10-3 80 78 88-10-4 73 72 88 -10-5 63 61

PAGE 56

Table 5-1-continued Carbonate Type CJ2-4-1 CJ2-4-2 CJ2-4-3 CJ2-4-4 CJ2-4-5 CJ8-1 0-1 CJS-10-2 CJ8-l 0-3 CJS-10-4 CJS-10-5 820-0-2 820-2-4 820-4-6 8206-8 820-8-10 CJ4-0-2 CJ4-2-4 CJ4-4-6 CJ4-6-8 CJ4-8-10 CJ-10-12 CJ-12-14 CJ-14-16 CJ-16-18 48 Total Mass Percent Carbonate 58 -t-0 . 4 0 81" 0 55 77 0 0 75 0 0 71 71 77 67 71 48 73 54 73 18 84 . ... Large Clasts From Two Centimeter Intervals Volume Percent Carbonate 55 0 . ' 3 0 80 0 53 75 0 0 74 0 0 69 69 76 65 69 46 71 51 71 17 83

PAGE 57

49 Volume% Carbonate 0 20 40 60 80 100 • 0 .. 2 I • J I 4 -E • 0 -6 • • •Core 820 0 (.) 8 I • Core CJ4 .c. I •• I (/) I I ::s 10 I I c: • :5 12 a. Q) 0 • 1 4 • 16 • 18 Figure 5-16 . Estimated carbonate volume percentage versus depth in core for two cores, measured from large , sieved clasts. The estimated porosity fluctuates with depth , and no clear patterns can be distinguished . Error bars are included on the two intervals with 6 data points.

PAGE 58

50 Figure 5-17. Transmitted-light photomicrograph of large acicular aragonite fans radiating from micritic aragonite. This photo taken from a chemoherm sample (box in Figure 7). Notice the large amounts of pore space which enables the growth of long aragonite needles.

PAGE 59

51 Figure 5-18 . Aragonite clasts cemented by aragonite needles observed in chemoherm and crust samples (transmitted light photomicrograph). These clasts are angular and poorly sorted , pointing to in situ cementation.

PAGE 60

52 Figure 51 9. Radiating fanl ike aragonite needles with blunt ends . SEM image taken from a chemoherm sample in the white aragonite zone . Notice t h e purity of the aragonite and the large pore spaces , supporting an orig i n above t h e seabed surface .

PAGE 61

53 Figure 5-20 . Transmitted light photomicrograph of a band in aragonite needle s in box from Figure 7. These structures seem to be a result of di s solution and reprecipitation , creating darker bands where less aragonite is pre sent.

PAGE 62

54 Fi g ure 5-21. SEM image of a band in aragonite needles from chemoherm s ample . The band formed during a time of dissolution and reprecipitation , w hich a ppear s a s a darker line along the needle s (black arrow) .

PAGE 63

55 Figure 5-22 . Small aragonite crystals growing from a micritic aragonite in a transmitted light photomicrograph of a crust. The change is fairly abrupt between the elongated needles and the sma ller micritic aragonite , possibly caused by a change in water chemistry and carbonate s uper sa turation. Micritic carbonate was determined to be aragonite and not high Mg calcite based on EDS samp ling that revealed Sr and no Mg.

PAGE 64

56 Figure 5-23. Boundary of aragonite matrix and clamshell (lower right) in an SEM photomicrograph. The aragonite is primarily micritic, but some radiating fan s are present. The boundary is poorly defined , and the aragonite is consistent in texture both proximal and distal from the clamshell .

PAGE 65

57 Figure 5-24 . Thin section photomicrograph of light tan and dark grey micritic aragonite commonly seen in chemoherm and crust samples (see Figure 8 for location). Whole and fragmented forams are commonly included in all the

PAGE 66

58 Figure 5-25 . SEM image of detrital material is common in the tan and grey aragonite cement . Clays and quartz are the most common detrital material included in the carbonates .

PAGE 67

59 Figure 5-26. Typical cement of a high Mg calcite sample seen in a transmitted light photomicrograph with broken forams , glauconite nodules , and low porosity.

PAGE 68

60 Figure 5-27. Example of micritic high Mg calcite in SEM photomicrograph. The crystals are all extremely small and poorly formed . This evidence supports an origin in an area with limited amount of room to grow (the sediment column).

PAGE 69

6 1 ' . .. ' ;'. . ; . . . . ' ... . .... .. ,_.... .. ... . . . . . . . . ; ..... . 1 .. ".;: .. .,.,..,.. p . er. ' .. . .. j' • ' . ' c._ • ... . • ._, , • • ••• ' " ....... • ... . . '<. . . ;,. . ' ' . ' • .r ... .,. • . • • " • • ... "" • . • • ,.. • . . 4oo; \ , < ' ' '"'"' . ' ' .. , '< -n , , c .... -.,, _ _., , -' . • . :. '-Nh'. " -'"""" . "" • ' f _ , .. -Q. • t. !. -_, •. ,...__._... ..... -.... . . • : . .,. , . • ••• ' . : 0 .... -' .... ; • . . . '...-..;, •'--';; ' . ... -;--:-.' ... . • . . • If' . • .., . \. . • .... , . • . . . .p • lo rr-. '..-J ,. , ..... , <.• < ' • ",.._' ,; • • •..... • e . , .. • .-, .. . • , . . .. • • • ' ; • .. • l . ' • • . • • .• . . . ''"' . .. . ' . ' . . ... ' ... . . . . .,' . ... • . . . . ;.,. . " . . . t • "' . . . .• ' . • . • . • .., .. . . . . . • . . • • < • • l" • J . ..... ' -. .,., • "' ..... ' -.L ,c;_. , _ '"..,.. . \ .,. ..... , ... . ,..... • . -_., , . • \ • . . . . . . . ' - •1': . ',., .••. :.. ;., • • • • I. . . • ... . I • \ " . ,. "'" " • •• . ' • - • ... .... . .. ... \ .. , .. / . { ' ,j \ ....... : .. • .. • .;:,J-."' '\(. ' .... . " . -. "'\ ,. . • . • ' ' < .. -... .. ... "'-"' . , . • • • • • I"' .. • ....... ' .. ' to• I ,-• • . • , !.' :{--.,.:..._ '#E .. J,, t. • ':!\ • -i . ,. • :,.. ' • . J. I • . .,.. . • .... (. -. f . . I ... . , • .... • • ' I •• ' ,. .. . : J . , • ' rr' .A ' . ' " ' ' • 4 ' ' V ' "-- • • . l ' • < \, ,_,. . • • • , ., \ f" • , . : > r-'" • 1 " • -ro 'r.• , .... , ; • .• . . : , .. • . . ... .._ :. , , . • ..._ .• , . "'. : .': .,. , --.,.,. . '{. '.. {\. ' . '\. ,. • ' 1 .. ' < . r.-';.: t . • > .... ; ,:. :. '._:,:• ' ,. !';9" ;.. v,. V..-V ( • • • ....,.. . l, ('. ..... , _ ' ... """--_...,._ .. "L:-"t... , \ .::.r .v"' . f. • • .. '. r ..,._ • !o._' ' .. t • ,h ..... '" ., -.,v = h . ,., , ... '•'' v' .•, . . ' _,. .. • . '""' ' . ' . ..... , ; . . , , . . ' 1;,., '"'-, IJ , , ...... . , '' , I ., . "" . . : "-.....:. . : . . .. . " . ;;/!!' , """ ( .. (_ ........ ..., ' . i.\_ I., _ ' • ' . .., ;....._ ' . . :!I. } • • . -. ,.(• '. < 1. .. P. ' • . ' . . . -. ' •... . . . -.. . \ ...._. . •, ' . A . • . 'V . • l " ' • ' • ,... -• • • • , . .., ;b • . > , , ' ' .,' ,.. . ,,.._ , • , ": • ! ;, • • • ; ;_.. . . 1': ' 1 < i.':]o_,,l . :<:;> • _ ; , l • ) • \ . .],f'. • , . , '-\; r • . . . • ' ' ' ... ..,.. t-• . • ' , . ' . . ' .. , ' • . i . ,:, : ' ,>.! .-c : , "i' ' ' I l . . . .... .... .. , ..... 1-. r .;"' \. : ... ._ ... . v . 'f f ' \ , .:...! . • • . , . . Figure 5 28 . Detrital material cemented by high Mg calcite seen in SEM . This quartz grain is cemented by poorly formed high Mg calcite.

PAGE 70

62 Figure 5-29 . This high Mg calcite sample (MS13) had the highest percentage of detrital material in any sample analyzed by SEM. Diatom frustules and clays are the most common detrital material in this image .

PAGE 71

63 Figure 5-30 . Banded micritic high Mg calcite and small clasts of pyrite, glauconite , and detrital material. The small clast zones include glauconite, and pyrite is present in both the micrite and the clast zones. This structure could be related to bands in the sediment during precipitation.

PAGE 72

6 4 Figure 5-31 . Dolomite sample with multiple veins. The micritic dolomite appears to be fractured and then reprecipitation occurs in the open spaces . N otice the very low percentage of pore space present in the thin section.

PAGE 73

65 Figure 5-32 . Dolomite crystals highl y magnified under SEM . The crystals are simi lar shapes and size , with some detrital material present.

PAGE 74

66 Figure 5-33. SEM image of the boundary between a dolomite clast (upper left) and a dolomite vein (lower right). The two matrices are indistinguishable, implying a simi lar precipitation environment for both. Notice the small percentage of pore space in the cement.

PAGE 75

67 Figure 5-34 . Transmitted-light photomicrograph of a representative breccia commonly found on Hydrate Ridge. The clasts are subangular, and comprised of high Mg calcite. The cement is acicular aragonite. High amounts of pore space allow for acicular aragonite growth and suggest an origin at or near the seabed surface . •

PAGE 76

68 Figure 5-35 . Breccia with large subrounded clasts , glauconite nodules , and shell fragments seen in transmitted-light photomicrograph . Aragonite needles cement the clasts and shells. Notice the large percentage of pore space .

PAGE 77

69 Figure 5-36. Transmitted-light photomicrograph of breccia with large clamshell, clasts , glauconite, and forams. Further support for an origin of breccias at the seabed surface i s the large whole forams present in thin section. These also support an in situ formation as opposed to a mass wasting event because the forams are unbroken .

PAGE 78

70 Figure 5-37. Pyrite infilling pore space in a high Mg calcite sample (transmitted-light photomicrograph) . These samples are much less porous and include more pyrite than the aragonitic samples .

PAGE 79

71 Figure 5-38. High Mg calcite sample with different colored vein seen in transmitted-light photomicrograph. The different colors are generally the same material with common forams and pyrite. Changes in color could be caused by differences in the amount of detrital material. Notice the low amount of pore space in the micritic cement, support ing precipitation in the sediment column. '.

PAGE 80

72 Figure 5-39. Edge ofhigh Mg calcite sample (located at the top of the photo) with large glauconite nodules as seen in transmitted-light photomicrograph. Dissolution of the glauconite from water flowing through the carbonate creates the brown hematite staining seen in this slide .

PAGE 81

73 Figure 5-40. SEM image offramboidal pyrite (as identified by SEM EDS) seen in a dolomite from Hydrate Ridge . Pyrite is common in dolomite and high Mg calcite samples , as seen in thin sections. •

PAGE 82

-55 -50 -45 iii 0 0.. Q. S: 0 C'J '0 -40 -35 -30 2 74 ... • .. • • ... • " • • X . ... --• • -,.: X -• • • • A c icul ar Aragonite • •Dark Grey Matrix • X .. • Chemoherm Dark Grey Matrix • x lightTan • xTan Grey Mi x • • X • Breccia cementa • • • • 2.5 3 3 . 5 4 4 . 5 5 5 . 5 6 6 .5 d180 (ppt PDB) Figure 5-41 . Stable isotopic values for aragonitic samples and breccias. Acicular aragonite has less positive 8180 values than other cements. Dark grey cement trends from more negative o13C and lower 8180 to less negative o13C and higher 8180. Chemoherm dark grey cement is relatively negative compared to other samples. The light grey and mixed cements have the highest 8180 values . The breccia cements are extremely random, with no obvious trends .

PAGE 83

-52.00 -4 7.00 -42 .00 -3 7.00 iii' -32 .00 Q a. ! 2 7.00 .. -a 22 .00 1 7.00 12 .00 -7.00 -2.00 2 . 00 75 • ... •r • • Pure Atagonite • Dal1< Aragoni t e . . . aom Shells . b . , .. r: '\ ' \ • I 2. 50 3 . 00 3. 50 4 .00 4 . 50 5 .00 5 . 50 6 .00 d180 (ppt PDB) Figure 5-42. Authigenic aragonite and clamshell isotopic compositions . The pure acicular aragonite has more depleted o 180 values than the darker grey aragonite, implying a greater influence of seawater during the pure aragoniti c precipitation. Noti ce that the grey and pure aragonite samples trend toward a common va l ue (near 4 ppt PDB) with more negat i ve o 13C values . The less negative o13C values for the pure and grey aragonite show a wide separat i o n b e tween o 1 80 values for each. Clamshells were plotted as a point of reference and as an example of typical isotop i c signatures of seawater at 4 C .

PAGE 84

76 0 3 . 0 3 . 5 4 . 0 4 . 5 5 . 0 5 . 5 6 . 0 d180 (ppt PDB) Figure 5-43 . Isotopic signatures for high Mg calcite samples . The o180 values are all positive, reflecting a complicated signature possibly associated with high temperature fluids during precipitation . The o13C reflects the influence of methane-rich waters during precipitation . The o 13C has a relatively wide range of values, which probably reflects ; a mixing of diagenetic marine water with biogenic methane waters . The isotopic values fall into three broad groups which reflect slightly itifluences during • e e • J . j' I t prectpttatlon. . .. . ... f) • • • : • " . " .. . . r.

PAGE 85

77 Tran s ect 1 Transect 2 del C-13 del 0-18 PDB PDB MDS086 Exterior -44 . 8 4.46 MDS087 -43.72 4.42 MDS088 -40.19 4.34 MDS089 41.89 4.43 MDS090 -34.93 3.66 MDS091 Interior -39.94 4.10 Figure 5-44. Shows table of high Mg calcite samples from cross sections for isotopic compositions . These samples were drilled across two calcite slabs (only one s lab is shown) . Transect 1 cross section depicts a clear trend in 813C values that decrease to the base of the slab. The 8180 values for the cross section are generally consistent , with a slight positive trend toward the base of the s lab. Transect 2 813C values generally become less negative toward the base of the slab . The 8180 values decrease slightly toward the base of the slab.

PAGE 86

iii Q a. Q, .e u .., ... , 78 , . ., .. • -30 • • .. • • 3 5 7 9 11 13 15 17 19 d1110 (ppt PDB) Figure 5-45. Isotopic signatures for dolomite samples. The o13C is less negative than values from aragonite or high Mg calcite samples. All 8180 values are positive, with the majority of the points falling between 5 and 7 ppt PDB, and one outlier at -41 %o PDB (not shown).

PAGE 87

CHAPTER6 DISCUSSION Published Models of Carbonate Precipitation One general model has been suggested for the precipitation of authigenic carbonates by several authors . Beauchamp et al . (1989) proposed that North Sea authigenic carbonates formed in the sulfate reduction zone. They also presented an in situ origin for the brecciation of carbonate crusts caused by high gas pressure (also causing the pockmarks seen in association with active venting). The pyrite found in most samples was attributed to occasional hydrogen sulfide seepage from deep in the sediment column (Beauchamp 1989) . All three mineralogies (aragonite, high Mg calcite, and dolomite) present in carbonates from the Gulf of Mexico were studied by Roberts and Aharon (1994). They offered a model of formation for all precipitates through bacteriallymediated reactions in the sulfate reduction zone or at the seabed surface (Roberts and Aharon 1994 ). Authigenic carbonates from the Florida escarpment are composed primarily of high Mg calcite, but aragonite and dolomite are present. Paull et al. ( 1992) suggested a general theory of formation similar to Roberts and Aharon ( 1994) where oxidation of methane and resulting sulfate reduction and increasing alkalinity promoted carbonate precipitation (Paull et al . 1994). These general models mentioned above provide an explanation for carbonate precipitation; however, they do not offer explanations for the presence of multiple mineralogies as seen on the Cascadia margin. Greinert et al. (2000) focused specifically on types and origin of the carbonates on Hydrate Ridge, and proposed precipitation in the near surface, slightly reducing conditions 79 . '

PAGE 88

80 in the top few centimeters or on the seabed , or from in situ brecciation of the mudclast and intraformational brecchias through slumping or venting . Aragonite precipitation requires near sediment-water interface conditions where high sulfate concentrations are present, while high Mg calcite precipitates where sulfate is reduced (sulfate reduction zone , >3 em depth). Calcite is the-favored phase at temperatures found on Hydrate Ridge , but the presence of sulfate greatly inhibits calcite precipitation and favors aragonite (see Literature Survey section) . Greinert et al. (2000) suggested that the dolomite mudstones originate at the deepest depths, in the lower methanogenic zone and below (10-220 mbst). They suggested this depth based on oxygen isotopic values that are highly negative (-3.5%o SMOW), and likely to occur at deeper depths where temperatures are higher (minimum 15.3C). Kulm et al. ( 1986) proposed a general origin for all carbonate cements at a shallow to surficial (upper 3 meters) depth based on <5180 values . Positive 0180 values indicate low temperature waters, which would occur at shallow depths (Kulm et al . 1986) . Behrmann et al . ( 1998) suggested that aragonite collected from chemoherms reflect recent seabed surface conditions, based on positive 0180 values (3 .68%o PDB) . However , they indicated that the high Mg calcite precipitated as a result of gas hydrate destabilization that occurred some time in the past, possibly caused by sea level lowering or an increase in temperature of the ocean. This conclusion was based on <5180 values which indicated high Mg calcite precipitated from porewater equilibrated with isotopically heavier (+3%o, from Behrmann et al. 1998) water derived from the release of cage waters during dissociation. The oxygen isotopic values for high Mg calcite in their study were 4 .86%o PDB which is enriched compared to average seawater ( +.95%o SMOW). Borhrnann et al.

PAGE 89

81 ( 1998) suggested that the 0180 enrichment was caused by the influence of hydrate-rich pore fluids (+3.0%o SMOW) . Previous investigators have promoted a variety of explanations for the different types of carbonates found on the Cascadia margin . Most papers agree that the carbonates are forming in the upper few tens of meters from the seafloor surface. However, they disagree on precisely where, and when the precipitation occurs . Morse et al. (1997) and Bohrmann et al . ( 1998) proposed a fluctuating system of high Mg calcite precipitation or aragonite precipitation based on climate and gas hydrate changes (Bohrmann et al. 1998, Morse et al . 1997). This explanation could only be applied to precipitation at the sediment surface because aragonite is unlikely to be the favored precipitate in the sulfate reduction zone due to the low temperatures and depleted sulfate (Burton and Walter 1987). A more likely explanation is high Mg calcite precipitating at depth while aragonite precipitates at the surface. The few authors that dealt with breccia formation on the Cascadia margin support a similar origin as presented herein , which suggests that gas hydrate destabilization causes either in situ brecciation or a massive gas discharge in the shallow seabed (Beauchamp et al . 1989, Greinert et al . 2000) . This theory is based on the presence of pockmarks, observations of floating gas hydrates, and analysis of the clasts and cements of the breccias (Bohrmann et al . 1998) . The dolomite is often believed to precipitate at greater depths, up to 40 mbsf (Greinert et al. 2000) . If precipitation occurs at this great of depth, exposure of the many large boulders on the seafloor surface of Hydrate Ridge would be difficult to explain through currents or gas hydrate destabilization which only involves the upper few meters of the seabed.

PAGE 90

82 Conceptual Model of Carbonate Precipitation The processes leading to the precipitation of carbonate phases on Hydrate Ridge are complicated, and involve multiple influencing factors including sulfate concentrations, temperatures, methane concentrations, and carbonate saturation state. The primary controls on formation of aragonite versus high Mg calcite promoted in this study are temperature of water during precipitation, porewater sulfate concentrations, and carbonate saturation state . The sulfate conditions are particularly important, because sulfate-rich or depleted zones can indicate depth of precipitation (Baker and Kastner 1981, Burton and Walter 1987) . Aragonite can only precipitate at the temperatures measured on Hydrate Ridge ( 4 C) if the precipitation occurs in the sulfate-rich sea waters, or if precipitation occurs in conjunction with warm-temperature migrating fluids seeping up from depth (Burton and Walter 1987). Even slightly higher than average temperatures (+0.32C) have only been recorded in a few places on the Cascadia margin because the high temperatures of migrating fluids are hard to maintain (Kulm et al . 1986). The warm waters quickly equilibrate with the surrounding colder porewaters through diffusion in a matter of days. Aragonite is usually found at the sediment water interface or in the water column as a chemoherm, and therefore the influence of sulfate is probably greater than migrating warm waters. The zone of sulfate reduction generally occurs below the upper few centimeters of the seabed, so the aragonite precipitation is not favored in this region (Greinert et al. 2000), and high Mg calcite, which is the favored phase at these temperatures, should precipitate in this zone (Figure 6-1 ). Dolomite precipitation is more difficult to understand where and how it occurs, or if the dolomite is authigenic (direct precipitation from seawater) or diagenetic (alteration

PAGE 91

83 from calcite) . Because sulfate greatly inhibits the dolomitization process, precipitation must occur somewhere in the sulfate reduction zone in the upper tens of meters from the seabed surface or below that where no sulfate is left. Baker and Kastner ( 1981) experimentally found that even minor amounts (0 .00 1 M) of sulfate greatly reduced the precipitation of dolomite. Minor sulfate has been found at depths as great at 60 mbsl on Hydrate Ridge (Westbrook et al . 1994). Baker and Bums (1985) also promoted a shallow sediment depth for dolomite precipitation, based on high magnesium and low sulfate porewater concentrations. Overall, the literature generally suggests that dolomitization can occur through replacement or authigenic precipitation, which can take place at depths up to a few tens of meters (Baker and Bums 1985, Baker and Kastner 1981). Aragonite Phase The thin section and SEM work are particularly important in understanding where mineralogies form . Aragonitic chemoherm samples have the largest amount of voids and the longest needles, which indicates high amounts of pore space and therefore precipitation at or above the seabed . They also have the largest percentage of carbonate (indicating the highest porosity), and a significant portion of pure aragonite (determined through XRD analysis). The presence of pure aragonite and absence of detrital material can only be achieved at or above the seabed surface. The different colored carbonates seen in chemoherm samples were composed of the same microcrystalline aragonitic and detrital mixture. This is important in establishing that no significant formational changes occurred to produce a mineralogy other than aragonite. Instead, the darker cement could indicate a higher percentage of detrital material, or reduced porosity due to dissolution and

PAGE 92

84 reprecipitation of microcrystalline aragonite (the percent carbonate analyses in this study were inconclusive) . Micritic aragonite could form by recrystallization even if precipitation occurs at or above the seabed where space is available for large crystals to grow. Reid and Macintyre ( 1998) suggested that micritic aragonite could precipitate in pore spaces or by recrystallization . The aragonitic needles seen in SEM and thin section seem to have banding, but on closer inspection the " bands" are gaps in the needles, probably indicative of periods of dissolution and then regrowth . SEM images also reveal the relative amounts of detrital material in the aragonite (less detrital) versus other phases . The low-detrital aragonite suggests precipitation at the seabed surface or above it. The highest percentage of whole forams is present in the aragonite samples, which indicates very little post-depositional compaction. All of these factors support a formation at or above the sediment water interface. One line of evidence that could support precipitation of aragonite at the sediment water interface is the absence of hematite staining and glauconite. If aragonite is forming at or slightly below the seafloor surface, then these samples should have the highest percentage of hematite staining due to the diffusion of seawater . Instead, the aragonite samples have almost no hematite stains present. This could imply a formation within a highly oxygenated zone, but recent precipitation has not allowed the oxidation of the pyrite and glauconite. The carbon isotopic data for the aragonite samples shows the strong influence of methane-rich waters during precipitation , implying that precipitation occurs near vents without too much mixing with seawater. The oxygen isotopic values of the aragonitic samples suggest the influence of seawater, with one sample averaging at a low 4.06%o PDB while the highest averaging at 4.89%o PDB. The o180 values of the

PAGE 93

85 formation water are calculated values of the 0180 during precipitation using standard equations discussed in the Methods section. Formation water o180 values from the chemoherm samples and the aragonitic samples ranges between -1.1 and O%o PDB and -0.3 and 0 .8%o PDB respectively (recording the lowest 0180 formation water values), implying precipitation influenced primarily by seawater (Grossman and Ku 1981). The modes of aragonite precipitation presented here are similar to the observations of others . Ritger et al. ( 1987) proposed a theory of aragonitic needles forming after the micritic calcite cement, which is only observed in this study amongst breccias . The micritic aragonite implies either recrystallization occurred or during precipitation there was little room for growth. The aragonite needles imply large open pore spaces and an origin above the seabed. The distinct boundary between the cement and needles could also point to a change in conditions during formation, or the inclusion of different amounts of detrital material (Ritger et al. 1987, this study). Bohrmann et al. (1998) and Greinert et al. (2000) proposed that the aragonitic needles grow into gas hydrate cavities. This theory could not be tested in this study, but is supported by the highly depleted carbon isotopic signature of the aragonite and the large cavities seen in thin section. Kulm et al. ( 1986) found well-sorted glauconite associated with the aragonite cement. However , the glauconite in the samples studied here is poorly sorted, and sometimes infrequent, and could indicate recent precipitation. This study also supports Ritger et al.' s ( 1987) correlation between higher induration and higher total carbonate percentage, which includes all three carbonate mineralogies . Ritger et al. ( 1987) noted two groups of samples, one with high mass percent CaC03 (25-30%) and the other with minimal carbonate (less than 20% CaC03).

PAGE 94

86 This bimodal pattern in Ritger's samples does not occur in the samples in this study, with the possible exception of clasts from the cores. Several clasts from cores contain practically no carbonate, while the remainders have a high CaC03 percentage (Figure 5-45) . Ritger' s association between high induration and high carbonate content is also recorded in this study, and is important in determining where the carbonate precipitated, as high detrital content reflects precipitation in the sediment column. Sample and Kopf ( 1995) reported much lower total carbonate percentages than Ritger or this study, with the highest values at 25% CaC03. Kulm et al. ( 1986) noted a difference in carbon and oxygen isotopic values of aragonitic cements versus aragonitic crusts. These trends are not seen in the data from this study, and could indicate a similar formation process for both the cements and the crusts (Kulm et al. 1986). Greinert et al. (2000) also recorded positive carbon isotopic values for chemoherm samples, where the chemoherm samples from this study center around -45%o PDB, with no outliers. Greinert et al. ' s positive carbon isotopic values could reflect the influence of seawater or higher temperatures on chemoherm samples, while this study's negative o13C values imply some influence of methane-rich waters. High Mg Calcite Phase High Mg calcite is favored for precipitation at 4-10 em depth in the zone of sulfate reduction. This depth of precipitation is based on the calculated porosity at the time of precipitation ( <10 em) and the location of the top of the sulfate reduction zone (>4 em). High Mg calcite is considered to be the dominant carbonate mineralogy present on the Cascadia margin, but some carbonates could have a high enough mol%Mg (>30

PAGE 95

87 mol%Mg) to be considered proto-dolomite . As stated previously, the Greinert classification system will be used in this paper since it is commonly used with carbonates from the Cascadia margin (high Mg calcite8-20 %; proto dolomite30-40%) . Ritger et al. (1987) had a bro a der range of values for high Mg calcite (6-20 mol%Mg) than this study (6-11 mol % Mg) , which could indicate a continual range of mol%Mg from high Mg calcite to dolomite . Sample and Reid ' s (1998) findings supported the minor constituents identified in this study in high-Mg calcites, which include clays , feldspars, and quartz. They identified the dominant cements as high Mg calcite and dolomite, partially concurring with this study . The same authors concluded that high Mg calcite was the primary cement, with a mol%Mg of 3 to 12. However, these conclusions were based on a small population size (35 samples) that was not sampled in a statistically random fashion (Sample and Reid 1998) . This zone of precipitation ( 4-10 em depth) is supported by reduced carbonate mass percen t ages (implying less porosity) , higher detrital material seen in thin s ection and SEM, and much less void space than aragonitic samples . Sediment porosities determined by gravimetric analyses of Alvin push core subsamples are 0.5-0.7 (50%-70% carbonate by mass) in the range of 4-10 em below the seafloor (Kastner pers . comm . ) . The presence of micrit i c calcite implies little room for crystal growth, and is cited as a support for high Mg calcite precipitation in the upper sediment column where little pore space inhibits crystal growth (Kopf et al. 1995) . A much higher percentage of high Mg calcite thin sections include broken forams than seen in aragonitic samples. The lack of hematite stains and glauconite in the high-Mg calcite samples supports a shallow burial, where water advects less than at the seafloor where water moves freely . Carbon isotopic values

PAGE 96

88 also show the importance of methane-rich waters during precipitation, with a range from -52%o PDB to -30%o PDB. The o180 values from high Mg calcite samples fall between 5 .18%o PDB and 4.13%o PDB . These very positive oxygen isotopic values reflect the importance of hydrate-rich waters during precipitation, and imply precipitation at some depth in the sediment column where seawater is not as influential. For example, the water in which the high Mg calcite formed has a calculated 01 80 value of +0. 7 to 2.7%o PDB at 3 . 9C (Tarutani et al . 1969). When methane hydrates dissociate, they release cage waters with o180 of 2 . 7 % o SMOW and therefore have a significant influence over the o180 values of the high Mg calcites . The high Mg calcite and dolomite observed in this study are similar to those described by given by Ritger et al . (1987) . The cement is primarily microcrystalline calcite and dolomite, with less pore space. Sample and Reid ( 1998) found that the calcite cement often included unbroken forams, bivalves, and veins, all of which are seen in this study . The veins have detrital grains accumulating in them, implying precipitation in the sediment column where grains can move into the veins . Most authors studying Cascadia carbonates . did not report any hematite staining in their thin sections, and dolomite descriptions and formation were rarely discussed. Our study found multiple hematite stains in a variety of carbonate types (high Mg calcite and dolomite) . This could indicate fluid flow and the presence of oxygen, and therefore whether the hematite formed at depth or at the seabed surface . Unfortunately , hematite staining could occur during formation or after exposure to the shallow sediment column, and cannot conclusively indicate depth of format i on .

PAGE 97

89 Ritger et al.' s ( 1987) group of high-Mg calcite samples all reflected a methane influence (highly depleted o13C values). Their values are similar to the ones from this study, except the range is larger and includes much lower values (as depleted as -66.7%o PDB, while this study's lowest sample is -53.2%o PDB). The Ritger et al. study also recorded relatively high oxygen isotopic values, similar to this study, with a range between +3.66 and +7.22%o PDB . The high oxygen isotopic values could be due to temperature changes or multiple events of hydrate formation and dissociation . Kulm and Suess ( 1990) also observed a broad range of negative o13C values in high Mg calcites (down to -66.7%o PDB) . The positive oxygen isotopic values seen in Ritger et al . , Kulm and Suess, and this study are probably indicative of a series of hydrate formation and dissociation , and dolomitization (Kastner, pers . comm., Kulm and Suess 1990, Sample and Reid 1998) . Dolomite Phase Dolomitic samples are common in this study , and are the dominant carbonate type in the cores . This result is surprising because authigenic dolomite precipitation is difficult to explain because experimental studies indicate that precipitation occurs in small amounts (Baker and Kastner 1981), and is not the dominant carbonate in Alvin-collected hand samples. Malone et al. ( 1996) suggested that authigenic dolomite in the Monterey Formation of California precipitated at times when migrating fluids in the shallow sediments reached temperatures over 1 00C. This theory could be applied to the Cascadia margin if periodic fluxes of warm fluids migrated to the seabed in the past 1-24 ky (dates from Kastner et al. 2000) . The possibility of large amounts of authigenic dolomite

PAGE 98

90 precipitating under modern conditions is difficult to explain because of colder temperatures; however, replacement dolomite could be a viable explanation of the large amounts of dolomite seen on Hydrate Ridge because high Mg calcite and aragonite will be replaced by dolomite . Unfortunately , replacement dolomite is impossible to distinguish from authigenic dolomite , and cannot be confirmed in this study . Because sulfate inhibits dolomiti z ation , dolomite must form in the zone of sulfate reduction (where pockets of complete sulfate depletion are present) or below it (where sulfate is not present) , in the upper tens of meters of the sediment column. Microenvironments of sulfate concentrations near zero can occur in the upper meters, and could be zones of dolomite precipitation . This study proposes a dolomite origin within the upper couple of meters of the sediment column, possibly at the same level as the high Mg calcite samples (>3 em depth) . This relatively shallow depth of formation is based on the amount of dolomite seen on the seabed surface and the availability of magnesium and calcium which is greatest in the upper couple of meters . Dolomites from Hydrate Ridge have an average of 61 mass percent CaC03 , which indicates less porosity than aragonite or high Mg calcite samples . The low fragmented forams, and micritic cement support precipitation in the sediment column (1-10 em depth). The presence of hematite staining and iron-manganese coating on many of the dolomite samples could attest to the old age of the dolomite and the movement of water through the dolomites since exposure to oxic seawater for a long period of time creates these rinds . Studies on iron-manganese crusts on the continental shelf off Tasmania suggest that these grow at rates of approximately 1 OJ.Lm/200, 000 years (Exon 1997) . The growth rates must be greater on the Cascadia margin since active

PAGE 99

91 venting has only occurred for the past 21-24 ky, but still point to an old age for the dolomite boulders (dates from Kastner et al . 2000) . The carbon isotope values from dolomitic samples range between -34.1%o and -39.6%o PDB, which reflect some influence of methane-rich waters . The o180 values for the dolomite fall within 5 .5%o and 7 .2%o PDB . Formation water o180 values of dolomite samples fall between 1.1 and 2 .8%o PDB , respectively . Sample and Reid (1998) reported isotopically similar dolomite samples to the ones in this study , as well as a group with slightly negative carbon isotope values (-1%o to -25%o) and negative oxygen isotope values (-4%o to -13%o). They attribute the differences in the isotopic values to changes in migrating fluids in different areas. This explanation could result in different samples with unique isotopic signatures dependent upon where the samples precipitated, which could explain some of the outlying carbon and oxygen isotopic values in their data and data presented in this study . Unfortunately, no systematic differences in the outliers exist and cannot directly support different locations of precipitation. Some aberrations exist within the data that does not fit with the model presented in this paper. These include no hematite staining in aragonite samples, and common hematite staining in dolomite samples . Also there are no clear trends in o13C volume percent carbonate values to suggest great influences of methane with depth . Finally , the presence of dolomite in all intervals of cores also does not fit the model of dolomite precipitating at depth as presented above.

PAGE 100

92 Breccias Authigenic carbonate breccias have been separated into a wide variety of categories by previous authors, including mudclast breccias, intraformational breccias, sandstones, and mudstones (Greinert et al. 2000 , Kopf et al. 1995). The breccias collected for this study primarily fall into mudclast and intraformational breccias . Intraformational breccias are the most common in this study, as many contain multiple types of clasts and shells , and are cemented by aragonite. The majority of the breccias analyzed are composed of detrital and high Mg calcite clasts with acicular aragonite cement. The breccias could be related to destabilized gas hydrates originating in the sediment column. The primary factors that support this conclusion come from observations in thin section and hand specimens. The breccias are composed of subangular to subrounded clasts, which are very poorly indurated . These observations are important because any transportation of these clasts would have rounded them or destroyed them completely. The matrix also commonly includes whole forams, which would have been crushed with any mass movement , such as a slump or turbidite flow . The high percentage of voids in breccia samples and the dominance of aragonitic cement in the breccias suggest cementation at or near the sediment water interface . Destabilized gas hydrates could break apart weakly cemented clasts, possibly sending material up into the water column and creating the pockmarks seen on Hydrate Ridge. The breccias seen on Hydrate Ridge could also have formed in situ through faulting or fracturing, which requires no transportation of clasts . Another possibility suggested by Suess et al. ( 1999) i s that floating hydrates can carry sediments and clasts with them and then release them as the hydrate dissociates . This material settles down to the seafloor, where i t is eventually

PAGE 101

93 cemented by aragonite. This theory can explain the angular shape of the clasts, variety of types , and the graded bedding occasionally seen in the breccias (Figure 5-9). .

PAGE 102

94 Large Chemohe rm Complexes Aragonite Sui tate rich waters Dolomite precipitation in the upper two meters r J Unstable Gas Hydrates 1 Methane-rich waters 1 migrating upwards Figure 6-1. Schematic of the ocean bottom and sediment column s howing precipitation areas of the carbonates. The aragonite forms at and above the seabed surface, where sulfate rich water promote precipitation in the form of chemoherm complexes and crusts. In the zone of sulfate reduction located below 4 em depth, Mg calcite is the favored precipitate. Dolomite is also inhibited by the presence of s ulfate, and therefore must precipitate in the sediment column, probably in the upper two meters. The gas hydrate stability zone is located at depths of approximately 100 meters. Pockmarks are also common on Hydrate Ridge.

PAGE 103

. CHAPTER 7 CONCLUSIONS The conclusions of this study include the following: • Three carbonate mineral phases are present on Hydrate Ridge and include aragonite , high-Mg calcite, and dolomite • The conditions under which these phases precipitate can be studied in thin section and SEM. These analyses reveal high porosity in aragonite samples with large acicular crystal growth, while the high-Mg calcite and dolomite samples reflect lower porosity and microcrystalline cement. Redox conditions are observed in thin section by the presence of pyrite and glauconite (reducing conditions), and hematite staining (oxidizing conditions). The aragonite and high-Mg calcite samples have little pyrite or glauconite and little to no hematite staining. The dolomite samples have high amounts of pyrite and hematite staining indicative of redox disequilibrium . • Stable isotopic analyses reflects the strong influence of methane-rich waters during all phases of carbonate precipitation in highly negative o13C values. The o180 values for aragonite samples are the closest to seawater values, implying a precipitation with the greatest influence of seawater. High Mg calcite and dolomite samples have very positive 0180 values, implying strong influence during precipitation of isotopically heavy cage waters released from gas hydrates • Each of the three mineral phases is precipitating primarily in response to the concentration of dissolved sulfate in the upper two meters of the seabed (aragonite at and above the seabed surface, high Mg calcite at 4-8 em depth, and dolomite at 4 em to 2 meters depth) • Oxidation of methane seeping from deep in the sediment column is the catalyst for carbonate production . Gas hydrate dissociation is a possible cause for breccia formation through gas hydrate destabilization The formation of authigenic carbonates off the coast of Cascadia is the product of methane released by tectonic and oceanographic processes. This type of methane-rich environment is more common on continental margins than previously believed, occurring in many oceans around the world. Future studies on authigenic, methane-related 95

PAGE 104

96 carbonates will be important to refine the myriad influences on precipitation, and will help to understand the interactions between gas hydrates and the carbonates. These studies could include an in-depth analysis of the dolomites present on Hydrate Ridge, radiometric age dating of the carbonates using UTh isotopes, more detailed study of the extent of carbonates, and experimental studies on carbonate precipitation under similar conditions as those occurring on Hydrate Ridge. A wide range of studies could be conducted on Hydrate Ridge which would help geologists to understand the formation of authigenic carbonates in modem ocean waters and in the rock record, and therefore help increase our understanding of tectonically-induced fluid flow on continental margins and the global carbon cycle.

PAGE 105

APPENDIX DATA FROM SAMPLES Table A-1. List of the total percent carbonate for drilled samples from carbonates, five single clasts from two intervals, and larger clasts from each interval. The clasts are from two cores, and intervals were taken every two centimeters. Number Name %CaC0J Type of Carbonate 6 mds002 71. 73 High Mg Calcite 7 mds003 62 .60 High Mg Calcite 8 mds004 68.76 High Mg Calcite 9 mds005 94 .44 High Mg Calcite 10 mds006 87.67 High Mg Calcite 11 mds007 72.67 High Mg Calcite 12 mdsOIO 54.32 High Mg Calcite 13 mdsOll 38 .99 High Mg Calcite 14 mds019 83.36 High Mg Calcite 15 mds021 68.70 High Mg Calcite 16 mds023 70.23 High Mg Calcite 18 mds025 83.64 High Mg Calcite 19 mds026 79 . 10 High Mg Calcite 20 mds027 84.41 High Mg Calcite 21 mds028 44. 08 High Mg Calcite 24 mds030 53.66 High Mg Calcite 25 mds031 57.95 High Mg Calcite 26 mds032 70.17 High Mg Calcite 27 mds034 73.21 High Mg Calcite 28 mds035 75 . 99 High Mg Calcite 29 mds038 72.99 Breccia 30 mdsOOl 57.49 Aragonite 31 mds039 78 . 54 Aragonite 32 mds042 86. 02 Aragonite 33 mds044 69.33 Aragonite 34 mds045 78.72 Aragonite 35 mds046 66.84 Aragonite 36 mds047 9l.l6 Aragonite 37 mds048 77.81 Aragonite 38 mfd049 76.30 Aragonite 39 mds050 69 . 55 Aragonite 40 mds049 76.67 Aragonite 41 mds026 84.83 Dolomite 42 mds028 44.62 Dolomite 97

PAGE 106

98 Core Samp l e Name %CaC03 Core Number Cor es 82-4-1 62 . 20 Single Cla s ts Core 1 82-4-3 69. 62 From Two Core 1 82-4-4 94 . 36 Interval s Core I 82-4-5 76. 34 Core I 88-J0-1 73 .31 Core I 8810-2 80 .51 Core I 8810-3 79 . 93 Core I 88-10-4 73 . 34 Core I 88-10-5 63 . 34 Core I CJ2-4-1 57 . 67 Core2 CJ2-4-2 0.40 Core2 CJ2-4-3 3 . 76 Core2 CJ2-4-4 0 .16 Core2 CJ2-4 5 81. 07 Core2 CJ8 -JO-I 0 .44 Core2 CJ8-I0-2 55 . 34 Core2 CJ8 J0-3 76. 84 Core2 CJ8-J0-4 0 .17 Core2 CJ8-10-5 0 . 30 Core2 820-0-2 75.41 Large C l a s ts Core 1 820-2-4 0 . 12 From Two Co r e I 820-4-6 0 .01 Centimeter Core I 820-6-8 70 . 90 I nterval s Core 1 8208-10 70 . 69 Core 1 CJ4-0-2 77 .33 Core2 CJ4 2-4 67 . 42 Core2 CJ4-4-6 70 . 99 Core2 CJ4-6-8 48.49 Core2 CJ4-8-JO 72 .99 Core2 CJ J0-12 53 . 65 Core2 CJ-12-14 73 . 24 Core2 CJ-14-16 18. 14 Core2 CJ1 6-18 84 . 39 Core2

PAGE 107

Sample# MDSOOl MDS002 MDS003 MDSOIO MDSOll MDS023 MDS024 MDS028 MDS029 MDS030 MDS031 MDS032 MDS034 MDS035 MDS038 MDS039 MDS042 MDS044 MDS045 MDS046 MDS047 MDS048 MDS049 MDS050 MDS051 MDS052 MDS053 MDS054 MDS055 MDS056 MDS057 MDS058 MDS059 MDS060 MDS061 MDS063 MDS064 MDS065 MDS066 MDS067 MDS068 MDS069 MDS070 99 Table A-2. Carbon and oxygen isotopic values for microdrilled carbonate samples from Hydrate Ridge. The formation water water o180 values are calculated as well. Microdrill del C-13 Sample PDB MDS001 MDS002 MDS003 MDS 010 MDS 011 MDS023 MDS024 MDS 028 MDS029 MDS 030 MDS 031 MDS032 MDS034 MDS035 MDS038 MDS 039 MDS042 MDS044 MDS045 MDS046 MDS047 MDS048 MDS049 MDS050 MDS 051 MDS052 MDS053 MDS054 MDS055 MDS056 MDS057 MDS058 MDS059 MDS060 MDS061 MDS063 MDS064 MDS065 MDS066 MDS067 MDS068 MDS069 MDS070 -53 . 2 -41.4 -42 . 2 -50.2 -51.4 -30.7 -35 . 0 -39.6 -37.6 -44.6 -44.6 -43.2 -47.8 -42.3 -40.0 -45.0 -47.4 -43.4 -43.7 -41.9 -42.1 -39.1 -42.3 -44.6 -45.22 -40.85 -45.28 -44.61 -46.19 -43.58 -45.2 -47.72 -47 . 68 -48.44 -47 .49 -49.91 -49 . 22 -45.03 -44.77 -46.04 -9.77 -5.58 -44.14 mean values del 0-18 of del 0-18 del 0-18 Water Mol % Big Delta formation PDB SMOW Temp(0C} MgC03 (SMOW) water SMOW 4.0 4 . 6 4.5 5 .7 5.1 5.4 5.2 5 .8 5.5 5.2 5 . 7 5.4 5.6 5 . 8 3.4 4.9 4.6 5.1 5.1 5 .2 5.1 4.9 5.0 5.0 3 . 92 3.81 3.88 3.88 3.8 3.99 3 . 83 4.01 4.4 4.42 4.29 4.12 4.15 4 . 8 4.8 4.64 4.41 4.64 4.78 35.0 35.6 35.6 36 . 7 36 . 2 36.4 36.2 36.9 36.6 36.3 36.8 36.5 36.7 36 .9 34.4 36.0 35.6 36.2 36.1 36.3 36.1 36. 0 36. 1 36.1 35.0 34.8 34.9 34.9 34.8 35.0 34.9 35.0 35.4 35.5 35.3 35.2 35.2 35.9 35.9 35.7 35.5 35.7 35.8 3.9 Aragonite 3.9 8 3.9 11 3.9 12 3.9 12 3.9 6 3.9 7 3.9 40 3.9 8 3 . 9 9 3.9 9 3.9 18 3.9 13 3.9 16 3.9 0 3.9 Aragonite 3 .9 Aragonite 3.9 Aragonite 3 .9 Aragonite 3.9 Aragonite 3.9 Aragonite 3.9 Aragonite 3 .9 Aragonite 3.9 Aragonite 3.9 Aragonite 3.9 Aragonite 3 .9 Aragonite 3.9 Aragonite 3.9 Aragonite 3 . 9 Aragonite 3.9 Aragonite 3.9 Aragonite 3.9 Aragonite 3.9 Aragonite 3.9 Aragonite 3.9 Aragonite 3.9 Aragonite 3.9 Aragonite 3.9 Aragonite 3.9 Aragonite 3.9 Clamshell 3.9 Clamshell 3 . 9 Aragonite 4 . 5 33.8 34.0 34.0 34.0 33.7 33.7 35.7 33 . 8 33.9 33.9 34.4 34.1 34 .3 33.3 4.5 4.5 4 . 5 4.5 4 . 5 4.5 4.5 4.5 4.5 4.5 4.5 4.5 4.5 4 . 5 4 . 5 4 . 5 4.5 4.5 4 . 5 4.5 4.5 4 . 5 4.5 4.5 4.5 4 . 5 4.5 4 . 5 -0.5 1.8 1.6 2.7 2.2 2.7 2.5 1.1 2.8 2.4 2.9 2.1 2.6 2.6 1.1 0.4 0.1 0.7 0.6 0.8 0 . 6 0.5 0 . 6 0.6 -0.5 -0.6 -0.6 -0.6 -0.7 -0.5 -0.6 -0.4 -0.1 0 . 0 -0.2 -0.3 -0.3 0.3 0.3 0 . 2 0 . 0 0.2 0.3

PAGE 108

Sample# MDS071 MDS072 MDS073 MDS074 MDS075 MDS076 MDS077 MDS078 MDS079 MDS080 MDS081 MDS082 MDS083 MDS084 MDS085 MDS086 MDS087 MDS088 MDS089 MDS090 MDS091 MDS092 MDS093 MDS094 MDS095 MDS096 MDS097 MDS098 MDS099 MDS100 MDS101 MDS102 MDS103 MDS104 MDS105 MDS106 MDS107 MDS108 MDS109 MDS110 MDS111 MDS112 MDS113 MDS114 MDS115 MDS116 MDS117 MDS118 Microdrill Sample MDS071 MDS072 MDS073 MDS074 MDS075 MDS076 MDS077 MDS078 MDS079 MDS080 MDS 081 MDS082 MDS083 MDS084 MDS085 MDS086 MDS087 MDS088 MDS089 MDS090 MDS091 MDS092 MDS093 del C-13 PDB -45:14 -43 . 94 -4.27 -0.48 -44.71 -44.77 -43.33 -30.69 -30 . 54 30.74 -30 . 62 -31.01 -31.44 -32 . 82 -36.19 -44.8 -43 . 72 -40 . 19 -41.89 -34 . 93 -39 . 94 -34.40 -39.77 MDS 094 did not run MDS 095 did not run MDS 096 -46 . 22 MDS097 MDS098 MDS099 MDS 100 MDS 101 MDS 102 MDS 103 MDS 104 MDS 105 MDS 106 MDS 107 MDS 108 MDS 109 MDS 110 MDS 111 MDS 112 MDS 113 MDS 114 MDS 115 MDS 116 MDS 117 MDS 118 -38.17 -39.53 -48.64 -1.16 -42.81 -46 . 25 -48 .51 15.01 -38.30 -35 . 97 -12.35 -36.37 -43 .68 -44.95 -42 . 13 -46.30 -51.57 -51.54 -51.61 -51.09 -44 .44 -47.84 100 mean vatue s del 0-18 del 0-18 PDB SMOW 4 . 73 35.8 4 . 95 36.0 4.49 35.5 4.4n 4.61 4 .61 4.9 5.36 5 . 03 5 . 29 5 . 05 5 .11 5 . 16 5.14 5.12 4.46 4.42 4 . 34 4.43 3 .66 4.10 3.59 3.01 4 . 77 5.31 5.20 5 .42 3 .51 6 . 29 5 . 66 5 . 98 4.45 3.45 3 . 22 2.84 3 . 96 5.07 6 . 24 5.27 6 . 57 5 .47 5 . 22 4.93 5 .41 3.72 3 . 77 35.5 35.7 35.7 36.0 36.4 36.1 36.4 36.1 36.2 36.2 36.2 36.2 35.5 35.5 35.4 35.5 34 . 7 35 . 1 34.6 34.0 35 . 8 36.4 36.3 36.5 34.5 37.4 36.7 37.1 35.5 34.5 34.2 33.8 35.0 36.1 37.3 36.3 37.7 36.5 36.3 36.0 36.5 34.7 34.8 Water Mol % Big Delta Temp(0C) 3.9 3 . 9 3.9 MgC03 (SMOW) Aragonite 4 . 5 Aragonite 4.5 Aragonite 4 . 5 3 . 9 Aragonite 3 . 9 Aragonite 3 . 9 Aragonite 3.9 Aragonite 3 . 9 7 3.9 7 3 . 9 7 3 . 9 7 3 . 9 7 3 . 9 7 3.9 7 3.9 7 3 . 9 11 3 . 9 11 3 . 9 II 3 . 9 11 3 . 9 11 3 . 9 11 3 . 9 Breccia 3 . 9 Breccia 3 . 9 3 . 9 3 . 9 3 . 9 3 . 9 3 . 9 3 . 9 3 . 9 3 . 9 3 . 9 3 . 9 3.9 3.9 3.9 3.9 3 . 9 3.9 3 . 9 3 . 9 3.9 3.9 3.9 3 . 9 3 . 9 3.9 Breccia Breccia 10 10 15 15. 15 13,46 13,46 13,46 13,46 Breccia Breccia Breccia Breccia Breccia Breccia Breccia Breccia 12 12 12 12 Aragonite Aragonite 4 . 5 4 . 5 4 . 5 4.5 33 . 7 33 . 7 33 . 7 33.7 33 . 7 33 . 7 33 . 7 33.7 34.0 34. 0 34.0 34 . 0 34 . 0 34 . 0 33 . 9 33 . 9 34.2 34.2 34 . 2 34.1 34.1 34.1 34. 1 34.0 34.0 34.0 34.0 4 . 5 4 . 5 del 0-18 of formation waterSMOW 0.3 0.5 0 . 0 0.0 0.2 0.2 0.4 2.7 2.3 2.6 2.4 2 . 4 2.5 2 . 5 2.4 1.5 1.5 1.4 1.5 0.7 1.1 1.9 2.5 2.0 2.3 0.3 3.3 2.6 3.0 1.4 2.5 2.2 1.9 2.4

PAGE 109

101 mean values del 0-18 of Microdrill del C-13 del 0-18 del 0-18 Water Mol% Big Delta formation Sample# Sample PDB PDB SMOW Temp(0C) MgC03 (SMOW) waterSMOW MDSII9 MDS 119 -43.53 3.73 34.8 3.9 Aragonite 4.5 MDSI20 MDS 120 -47.47 3 . 83 34.9 3.9 Aragonite 4.5 MDSI21 MDS 121 -46.63 3.83 34.9 3.9 Aragonite 4.5 MDSI22 MDS 122 -45.60 3.65 34.7 3.9 Aragonite 4.5 MDSI23 MDS 123 -42.05 4.68 35.7 3.9 Breccia MDSI24 MDS 124 -39.43 4.52 35.6 3 . 9 Breccia MDSI25 MDS 125 -40.22 4.72 35.8 3.9 Breccia MDSI26 MDS 126 -38.36 4.74 35. 8 3.9 Breccia MDSI27 MDS 127 -3 5.49 3.38 34.4 3.9 Breccia MDSI28 MDS 128 -36.21 3.30 34.3 3.9 Breccia C2 C2 -46.71 14.01 34.7 Aragonite C3 C3 -49.68 14.10 34.8 Aragonite C5 C5 -2.44 14.87 35.5 Clamshell C7 C7 -46.55 13.86 34.5 Aragonite C8 C8 -49.00 14.59 35.2 Aragonite C9 C9 -46.24 14.70 35.4 Aragonite ClO CIO 53.87 14.00 34.6 Aragonite 5.7 -2.0 Cll Cll -54.65 15.34 36.0 3.9 8 33.8 2.2 MS07 Cl2 -35.72 17.45 38.2 4 42 35.8 2.3 MSll Cl3 -32.26 15.33 36. 0 4 11 34.0 2.0 MSil Cl4 -42.45 15.29 36 . 0 4 7 33.7 2.2 MS05 Cl5 -31.75 15.52 36.2 4 6 33.7 2.5 MSI7 Cl6 -22.5 1 1 3.32 33.9 4 8 33.8 0.2 MSI7 Cl7 -38.95 13.23 33.9 9 2.9 -0.1 MSI3 CIS -46.72 15.57 36.3 4 14 34.1 2.1 MSI3 Cl9 43.49 14 .67 35.3 4 4.4 -0.1 MS15 C20 -44.86 14.94 35.6 4 13 34.1 1.5 MS15 C21 -44.51 14.73 35.4 4 4.4 -0.1

PAGE 110

LIST OF REFERENCES Adshead, J . 1996. Stable Isotopes, 14C Dating, and Geochemical Characteristics of Carbonate Nodules and Sediment from an Active Vent Field, Northern Juan De Fuca Ridge, Northeast Pacific. Chemical Geology, v . 129. 133-152 . Aharon, P . 1994. Geology and Biology of Modem and Ancient Submarine Hydrocarbon Seeps and Vents: an Introduction . Geo-Marine Letters, v. 14. 69-73. Baker , P., S . Bums. 1985. Occurrence and Formation of Dolomite in Organic-Rich Continental Margin Sediments . The American Association of Petroleum Geologists Bulletin, v. 69. 1917-1930 . Baker, P ., M . Kastner . 1981. Constraints on the Formation of Sedimentary Dolomite. Science, v . 213. 214-216. Beauchamp, B., H . Krouse, J . Harrison, W . Nassichuk, L. Eluik . 1989. Cretaceous Cold Seep Communities and Methane-Derived Carbonates in the Canadian Arctic . Science, v. 244. 53-56. Beauchamp, B., M. Savard. 1992. Cretaceous Chemosynthetic Carbonate Mounds in the Canadian Arctic. Society for Sedimentary Geology, v. 7. 434-450. Berner, R., 1980. Early Diagenesisa Theoretical Approach. Princeton University Press, Princeton, N.J. Bohrmann, G., J. Greinert, E. Suess, M . Torres . 1998 . Authigenic Carbonates from the Cascadia Subduction Zone and Their Relation to Gas Hydrate Stability . Geology, v. 26. 647-650 . Burton, E., L. Walter. 1987. Relative Precipitation Rates of Aragonite and Mg Calcite from Seawater: Temperature or Carbonate Ion Control? Geology, v . 15 . 111-114. Campbell, K. A. 1992. Recognition of a Mio-Pliocene Cold Seep Setting from the Northeast Pacific Convergent Margin, Washington, USA. Society for Sedimentary Geology, v. 7. 422-433. Carson, B., M. Holmes, K . Umstattd, J. Strasser, H. Johnson. 1991. Fluid Expulsion from the Cascadia Accretionary Prism: Evidence From Porosity Distribution, Direct Measurements, and GLORIA Imagery. Phil. Trans. R. Soc. London, v. 335 . 331-340. 102

PAGE 111

103 Carson, B . , E . Screaton . 1998. Fluid Flow in Accretionary Prisms: Evidence for Focused, Time-Variable Discharge . Reviews of Geophysics, v. 36. 329-351. Carson, B . , E . Seke, V . Paskevich, M. Holmes. 1994. Fluid Expulsion Sites on the Cascadia Accretionary Prism : Mapping Diagenetic Deposits with Processed GLORIA Imagery. Journal of Geophysical Research , v . 99. 11,959-11969 . Devol, A . , J . Anderson, K . Kuivila, J. Murray. 1984. a Model for Coupled Sulfate Reduction and Methane Oxidation in the Sediments of Saanich Inlet. Geochimica et Cosmochimica Acta, v . 48. 993-1004 . Eichhubl, P., J. Boles . 2000 . Rates of Fluid Flow in Fault SystemsEvidence for Episodic Rapid Fluid Flow in the Miocene Monterey Formation, Coastal California. American Journal of Science , v. 300. 571-600. Exon, N. 1997. Ferromanganese Crust and Nodule Deposits from the Continental Margin South and West of Tasmania. Australian Journal of Earth Sc i ences, v. 44. 701-710 . Friedman, G . Methane-Derived Authigenic Carbonates Formed by Subduction-Induced Pore-Water Expulsion Along the Oregon/Washington Margin : Discussion and Reply . 1988 . Geological Society of America Bulletin, v. 100 . 622-623. Friedman , I., J. O'Neil. 1977 . Compilation of Stable Isotope Fractionation Factors of Geochemical Interest. Data of Geochemistry. United States Government Printing Office, Washington, D . C. Goldsmith , J., D . Graf , H . Heard . 1961. Lattice Constants of the Calcium-Magnesium Carbonates. American Mineralogist, v . 46. 453-457. Greinert, J ., G . Bohrmann, E . Suess . 2000 . Gas Hydrate-Associated Carbonates and MethaneVenting at Hydrate Ridge : Classification, Distribution, and Origin of Authigenic Lithologies . Geophysical Monograph , v . 124 . 99-113. Grossman, E., T . Ku. 1981. Aragonite-Water Isotopic Paleotemperature Scale Based on the Benthic Foraminifer Hoeglundina Elegans . Abstracts with ProgramsGeological Society of America, v. 13. 464-468. Hudson, J., T . Anderson . 1989 . Ocean Temperatures ar:td Isotopic Compositions through Time . Transactions of the Royal Society of Edinburgh : Earth Sciences, v . 80. 183-192. Hyndman, R., G. Spence, T . Yuan , E. Davis. 1995. Gas Hydrate Studies on the West Coast of Vancouver Island . Ocean Drilling Program, Initial Reports, v. 146 . 399-419. Jorgensen, N. 1990 . Methane-Derived Carbonate Cementation of Holocene Marine Sediments from Kattegat, Denmark. Continental Shelf Research, v . 12. 1209-1218 .

PAGE 112

104 Kastner, M ., P . Baker . 1.981. Constraints on the Formation of Sedimentary Dolomite. Science, v . 213 . 214-216. Kastner, M. , B . Carson, D. B . Bartlett, J. Jaeger , K . Bidle, and H. Jannasch. 2000. an in Situ Experiment of Methane Sequestration as Gas Hydrate , Authigenic Carbonate, and Loss to the Water Column and/or Atmosphere . Gas Hydrates : Challenges for the Future, G . D. Holder and P.R. Bishnoi, Ed. New York, New York . 32-37 . Kastner, M . , K . Kvenvolden , T . Lorenson . 1998 . Chemistry, Isotopic Composition, and Origin of a Methane-Hydrogen Sulfide Hydrate at the Cascadia Subduction Zone. Earth and Planetary Science Letters, v. 156. 173-183 . Kopf , A., J. Sample , P. Bauer, J . Behrmann, H . Erlenkeuser. 1995. Diagenetic Carbonates from Cascadia Margin : Textures , Chemicals, Compositions, and Oxygen and Carbon Stable Isotope Signatures . Proceedings of the Ocean Drilling Program, Scientific Results, v . 146 (Pt 1) . 117-136 . Kulm, L . D . , E . Suess. 1990. Relationship between Carbonate Deposits and Fluid Venting : Oregon Accretionary Prism. Journal of Geophysical Research , Special Issue : Fluids in Accretionary Prisms . Kulm, L . D., E. Suess , J . C . Moore , B . Carson, B.T . Lewis , S . D . Ritger, D . C . Kadko, T.M . Thornburg, R . W. Embley, W . D. Rugh, G.J . Massoth , M . G . Langseth, G . R . Cochrane, R.L. Scamman. 1986 . Oregon Subduction Zone : Venting, Fauna, and Carbonates. Science , v. 231 . 561-566 . Kvenvolden, K. 1995. a Review of the Geochemistry of Methane in Natural Gas Hydrate. Organic Chemistry, v . 23 . 997-1008. Malone, M. , P . Baker, S. Burns. 1996. Hydrothermal Dolomitization and Recrystallization of Dolomite Breccias from the Miocene Monterey Formation, Tepusquet Area, California . Journal of Sedimentary Research , v. 66. 976 990. Martin, J. B ., M . Kastner, P . Henry , X . Le Pichon , S . Lallement. 1996. Chemical and Isotopic Evidence for Sources of Fluids in a Mud Volcano Field Seaward of the Barbados Accretionary Wedge. Journal of Geophysical Research, v . 101. 20,325-20,245 . Matsumoto, R . 1989 . Isotopically Heavy Oxygen-Containing Siderite Derived from the Decomposition of Methane Hydrate. Geology, v. 17 . 707-710. Morse, J ., Q. Wang, M. Tsio. 1997. Influences of Temperature and Mg:Ca Ratio on CaC03 Precipitates from Seawater . Geology, v. 25 . 85-87. Paull, C . K . , J.P. Chanton , A . C. Neumann, J . A. Coston, C . S . Martens, W. Showers. 1992 . Indicators of Methane-Derived Carbonates and Chemosynthetic Organic Carbon

PAGE 113

105 Depo s its : Examples from the Florida Escarpment. Society for Sedimentary Geology, v . 7 . 361-375 . Peckmann, J . , A. Reimer, U . Luth, C. Luth, B. Hansen, C. Heinicke , J . Hoefs , J . Reitner . 2001 . Methane-Derived Carbonates and Authigenic Pyrite from the Northwestern Black Sea . Marine Geology, v . 177 . 129-150. Rad, U., H . Rosch, U. Berner, M. Geyh, V. Marchig, H . Schulz . 1996 . Authigenic Carbonates Derived from Oxidized Methane Vented from the Makran Accretionary Prism off Pakistan . Marine Geology, v . 136. 55-77. Reid, R., I . Macintyre. 1998. Carbonate Recrystallization in Shallow Marine Environments : a Widespread Diagenetic Process Forming Micritized Grains . Journal of Sedimentary Research, v . 68 . 928-946 . Ritger, S., B . Carson, E. Seuss. 1987. Methane-Derived Authigenic Carbonates Formed by SubductionInduced Pore Water Expulsion Along the Oregon!W ashington Margin. Geological Society of America Bulletin, v. 98 . 147-156 . Roberts, H . , P. Aharon. 1994. Hydrocarbon-Derived Carbonate Buildups of the Northern Gulf of Mexico Continental Slope: a Review of Submersible Invest i gations . Geo-Marine Letters, v. 14 . 135-148 . Sample, J . 1996 . Isotopic Evidence from Authigenic Carbonates for Rapid Upward Fluid Flow in Accretionary Wedges . Geology, v . 24. 897-900 . Sample, J., A . Kopf . 1995. Isotope Geochemistry of Syntectonic Carbonate Cements and Veins from the Oregon Margin : Implications for the Hydrogeologic Evolution of the Accretionary Wedge. Proceedings of the Ocean Drilling Program, Scientific Results, v. 146 (Pt 1). 137-148. Sample, J ., M. Reid. 1998. Contrasting Hydrogeologic Regimes Along Strike-Slip and Thrust Faults in the Oregon Convergent Margin: Evidence from the Chemistry of Syntectonic Carbonate Cements and Veins. GSA Bulletin, v . 110 . 48-59. Sample, J., M. Reid, H . Tobin , J. Moore . 1993. Carbonate Cements Indicate Channeled Fluid Flow along a Zone of Vertical Faults at the Deformation Front of the Cascadia Accretionary Wedge (Northwest U . S. Coast). Geology. 507-510 . Scamman, R. 1981 . Diagenetic Carbonate Cementation of Clastic Sediments Near the Sediment-Water Interface on the Lower Continental Slope off Washington and Northern Oregon (M.S . thesis) . Bethlehem, Pennsylvania, Lehigh University . 1-197 . Schluter, M . , P . Linke, E . Suess . 1998. Geochemistry of a Sealed Deep-Sea Borehole on the Cascadia Margin . Marine Geology, v. 148 . 9-20 .

PAGE 114

106 Suess, E., M. Torres, G . Bohrmann, R. Collier, J. Greinert, P. Linke, G. Rehder, A. Trehu, K. Wallmann, G. Winckler, E. Zuleger. 1999. Gas Hydrate Destabilization: Enhanced Dewatering, Benthic Material Turnover and Large Methane Plumes at the Cascadia Convergent Margin. Earth and Planetary Science Letters, v . 170. 1-15. Suess, E., G. Bohnnann, R. von Huene, P. Linke, K. Wallman, S. Lammers, H . Sahling. 1998. Fluid Venting in the Eastern Aleutian Subduction Zone. Journal of Geophysical Research , v . 103 . 2597-2614. Tarutani, T., R. Clayton, T. Mayeda. 1969. The Effect of Polymorphism and Magnesium Substitution on Oxygen Isotope Fractionation between Calcium Carbonate and Water. Geochimica et Cosmochimica Acta, v. 33. 987-996. Teal, C., S. Mazzulo, W. Bischoff. 2000. Dolomitization of Holocene Shallow-Marine Deposits Mediated by Sulfate Reduction and Methanogenesis in Normal-Salinity Seawater, Northern Belize. Journal of Sedimentary Research, v. 70. 649-663. Terzi, C., P. Aharon, F. R. Lucchi, G. B. V ai.1994. Petrography and Stable Isotope Aspects of ColdVent Activity Imprinted on Miocene-age "Calcari a Lucina" from Tuscan and Romagna Apennines, Italy. Geo-Marine Letters, v. 14 . 177-184. Trehu, A . , M. Torres, G . Moore, E. Suess, G. Bohrmann. 1999. Temporal and Spatial Evolution of a Gas Hydrate-Bearing Accretionary Ridge on the Oregon Continental Margin. Geology, v. 27. 939-942 . Vasconcelos, C., J. McKenzie. 1997. Microbial Mediation of Modern Dolomite Precipitation and Diagenesis under Anoxic Conditions (Lagoa V ermelha, Rio de Janeiro, Brazil). Journal of Sedimentary Research, v. 67 . 378-390 . Westbrook, G., B. Carson, R. Musgrave. 1994. Proceedings of the Ocean Drilling Program, Initial Reports, v. 146 (part 1). 301-375. Whiticar, M., E. Faber. 1986 . Methane Oxidation in Sediment and Water Column EnvironmentsIsotopic Evidence. Organic Geochemistry, v. 10. 759-768.

PAGE 115

BIOGRAPHICAL SKETCH Mary Lindsey Bateman was born in Memphis, Tennessee to Randolf and Victoria Bateman . She was raised in Austin, Texas, and graduated from Indian Hill High School in Cincinnati Ohio in 1995 . She then graduated magna cum laude and Phi Beta Kappa with a BA degree from Southern Methodist University in Dallas , Texas, where she majored in both geology and anthropology . Mary Lindsey continued her education at the University of Florida, where she graduated with an MS degree in 2002. She presently resides with her dogs Isabeau and Clovis in Houston , where she works as a petroleum geologist. 107

PAGE 116

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a thesis for the degree of Master of Science . of Geological Sciences I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a thesis for the degree of Master of Science . '")-. 'JL ... L . ., Anthony Randazzo } Professor of Geological Sciences I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a thesis for the degree of Master of Science. (JIM I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a thesis for the degree of Master of Science. r Associate Professor of Anthropology This thesis was submitted to the Graduate Faculty of the College of Liberal Arts and Sciences and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Master of Science. December 2002 Winfred Phillips Dean, Graduate School

PAGE 117

3 1262 07299 099 6


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID ECFSUOTKD_VJIPPE INGEST_TIME 2015-04-20T20:59:54Z PACKAGE AA00030918_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES